id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
sequence |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.02353 | Zhanxin Wu | Zhanxin Wu, Bo Ai, Tom Silver, Tapomayukh Bhattacharjee | SAVOR: Skill Affordance Learning from Visuo-Haptic Perception for Robot-Assisted Bite Acquisition | null | null | null | null | cs.RO | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Robot-assisted feeding requires reliable bite acquisition, a challenging task due to the complex interactions between utensils and food with diverse physical properties. These interactions are further complicated by the temporal variability of food properties-for example, steak becomes firm as it cools even during a meal. To address this, we propose SAVOR, a novel approach for learning skill affordances for bite acquisition-how suitable a manipulation skill (e.g., skewering, scooping) is for a given utensil-food interaction. In our formulation, skill affordances arise from the combination of tool affordances (what a utensil can do) and food affordances (what the food allows). Tool affordances are learned offline through calibration, where different utensils interact with a variety of foods to model their functional capabilities. Food affordances are characterized by physical properties such as softness, moisture, and viscosity, initially inferred through commonsense reasoning using a visually-conditioned language model and then dynamically refined through online multi-modal visuo-haptic perception using SAVOR-Net during interaction. Our method integrates these offline and online estimates to predict skill affordances in real time, enabling the robot to select the most appropriate skill for each food item. Evaluated on 20 single-item foods and 10 in-the-wild meals, our approach improves bite acquisition success by 13% over state-of-the-art (SOTA) category-based methods (e.g. use skewer for fruits). These results highlight the importance of modeling interaction-driven skill affordances for generalizable and effective robot-assisted bite acquisition. Website: https://emprise.cs.cornell.edu/savor/ | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:14:45 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wu",
"Zhanxin",
""
],
[
"Ai",
"Bo",
""
],
[
"Silver",
"Tom",
""
],
[
"Bhattacharjee",
"Tapomayukh",
""
]
] |
2506.02354 | Jianzong Wang | Junjie Li, Nan Zhang, Xiaoyang Qu, Kai Lu, Guokuan Li, Jiguang Wan, Jianzong Wang | RATE-Nav: Region-Aware Termination Enhancement for Zero-shot Object Navigation with Vision-Language Models | Accepted by the 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025) | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Object Navigation (ObjectNav) is a fundamental task in embodied artificial intelligence. Although significant progress has been made in semantic map construction and target direction prediction in current research, redundant exploration and exploration failures remain inevitable. A critical but underexplored direction is the timely termination of exploration to overcome these challenges. We observe a diminishing marginal effect between exploration steps and exploration rates and analyze the cost-benefit relationship of exploration. Inspired by this, we propose RATE-Nav, a Region-Aware Termination-Enhanced method. It includes a geometric predictive region segmentation algorithm and region-Based exploration estimation algorithm for exploration rate calculation. By leveraging the visual question answering capabilities of visual language models (VLMs) and exploration rates enables efficient termination.RATE-Nav achieves a success rate of 67.8% and an SPL of 31.3% on the HM3D dataset. And on the more challenging MP3D dataset, RATE-Nav shows approximately 10% improvement over previous zero-shot methods. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:15:00 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Junjie",
""
],
[
"Zhang",
"Nan",
""
],
[
"Qu",
"Xiaoyang",
""
],
[
"Lu",
"Kai",
""
],
[
"Li",
"Guokuan",
""
],
[
"Wan",
"Jiguang",
""
],
[
"Wang",
"Jianzong",
""
]
] |
2506.02355 | Andre He | Andre He, Daniel Fried, Sean Welleck | Rewarding the Unlikely: Lifting GRPO Beyond Distribution Sharpening | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Reinforcement learning has emerged as an effective framework for training large language models on structured language-conditioned tasks. We identify a critical flaw of Group Relative Policy Optimization (GRPO), a widely used RL algorithm in this setting. For tasks that require multi-sample performance, such as formal theorem proving, GRPO biasedly reinforces already probable solutions and neglects rare but correct proofs. This implicit bias impairs performance on pass@$N$ metrics at large sample sizes, limiting its practicality for training theorem provers. To address this, we introduce the unlikeliness reward, a straightforward method that explicitly encourages reinforcing rare correct solutions. Additionally, we find that increasing the number of PPO epochs further mitigates this bias. Our experiments confirm that incorporating the unlikeliness reward significantly improves pass@$N$ across a large range of N, outperforming standard GRPO and substantially increasing sample diversity. Applying our revised recipe to Lean, we achieve competitive performance with DeepSeek-Prover-V1.5-RL on the miniF2F-test benchmark. We release our implementation, providing a simple yet effective recipe for training formal theorem provers with RL. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:15:15 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"He",
"Andre",
""
],
[
"Fried",
"Daniel",
""
],
[
"Welleck",
"Sean",
""
]
] |
2506.02357 | Ram Potham | Ram Potham (Independent Researcher) | Evaluating LLM Agent Adherence to Hierarchical Safety Principles: A Lightweight Benchmark for Probing Foundational Controllability Components | Preprint. This work has been submitted to the Technical AI Governance Workshop at ICML 2025 for review | null | null | null | cs.LG cs.AI cs.CY | http://creativecommons.org/licenses/by/4.0/ | Credible safety plans for advanced AI development require methods to verify agent behavior and detect potential control deficiencies early. A fundamental aspect is ensuring agents adhere to safety-critical principles, especially when these conflict with operational goals. Failure to prioritize such principles indicates a potential basic control failure. This paper introduces a lightweight, interpretable benchmark methodology using a simple grid world to evaluate an LLM agent's ability to uphold a predefined, high-level safety principle (e.g., "never enter hazardous zones") when faced with conflicting lower-level task instructions. We probe whether the agent reliably prioritizes the inviolable directive, testing a foundational controllability aspect of LLMs. This pilot study demonstrates the methodology's feasibility, offers preliminary insights into agent behavior under principle conflict, and discusses how such benchmarks can contribute empirical evidence for assessing controllability. We argue that evaluating adherence to hierarchical principles is a crucial early step in understanding our capacity to build governable AI systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:16:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Potham",
"Ram",
"",
"Independent Researcher"
]
] |
2506.02358 | Tianze Wang | Tianze Wang, Zhang Zhang, Chao Sun | RoadFormer : Local-Global Feature Fusion for Road Surface Classification in Autonomous Driving | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The classification of the type of road surface (RSC) aims to utilize pavement features to identify the roughness, wet and dry conditions, and material information of the road surface. Due to its ability to effectively enhance road safety and traffic management, it has received widespread attention in recent years. In autonomous driving, accurate RSC allows vehicles to better understand the road environment, adjust driving strategies, and ensure a safer and more efficient driving experience. For a long time, vision-based RSC has been favored. However, existing visual classification methods have overlooked the exploration of fine-grained classification of pavement types (such as similar pavement textures). In this work, we propose a pure vision-based fine-grained RSC method for autonomous driving scenarios, which fuses local and global feature information through the stacking of convolutional and transformer modules. We further explore the stacking strategies of local and global feature extraction modules to find the optimal feature extraction strategy. In addition, since fine-grained tasks also face the challenge of relatively large intra-class differences and relatively small inter-class differences, we propose a Foreground-Background Module (FBM) that effectively extracts fine-grained context features of the pavement, enhancing the classification ability for complex pavements. Experiments conducted on a large-scale pavement dataset containing one million samples and a simplified dataset reorganized from this dataset achieved Top-1 classification accuracies of 92.52% and 96.50%, respectively, improving by 5.69% to 12.84% compared to SOTA methods. These results demonstrate that RoadFormer outperforms existing methods in RSC tasks, providing significant progress in improving the reliability of pavement perception in autonomous driving systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:23:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wang",
"Tianze",
""
],
[
"Zhang",
"Zhang",
""
],
[
"Sun",
"Chao",
""
]
] |
2506.02359 | Brent Griffin Dr | Brent A. Griffin, Manushree Gangwar, Jacob Sela, Jason J. Corso | Auto-Labeling Data for Object Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Great labels make great models. However, traditional labeling approaches for tasks like object detection have substantial costs at scale. Furthermore, alternatives to fully-supervised object detection either lose functionality or require larger models with prohibitive computational costs for inference at scale. To that end, this paper addresses the problem of training standard object detection models without any ground truth labels. Instead, we configure previously-trained vision-language foundation models to generate application-specific pseudo "ground truth" labels. These auto-generated labels directly integrate with existing model training frameworks, and we subsequently train lightweight detection models that are computationally efficient. In this way, we avoid the costs of traditional labeling, leverage the knowledge of vision-language models, and keep the efficiency of lightweight models for practical application. We perform exhaustive experiments across multiple labeling configurations, downstream inference models, and datasets to establish best practices and set an extensive auto-labeling benchmark. From our results, we find that our approach is a viable alternative to standard labeling in that it maintains competitive performance on multiple datasets and substantially reduces labeling time and costs. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:27:56 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Griffin",
"Brent A.",
""
],
[
"Gangwar",
"Manushree",
""
],
[
"Sela",
"Jacob",
""
],
[
"Corso",
"Jason J.",
""
]
] |
2506.02360 | Yoshinori Hashimoto | Yoshinori Hashimoto | Relative uniform $K$-stability over models implies existence of extremal metrics | 12 pages | null | null | null | math.DG math.AG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We prove that an extremal metric on a polarised smooth complex projective variety exists if it is $\mathbb{G}$-uniformly $K$-stable relative to the extremal torus over models, extending a result due to Chi Li for constant scalar curvature K\"ahler metrics. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:31:29 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Hashimoto",
"Yoshinori",
""
]
] |
2506.02361 | Chang Li | Chang Li | Spin Grouping in Ring Cavity and its Protection on Entangled States Transfer | 7+5 pages, 4+4 figures | Phys. Rev. Lett. 134, 210803 (2025) | null | null | quant-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Long-range interactions are essential for large-scale quantum computation and quantum interconnections. Cavities provide a promising avenue to achieve long-range interaction by enhancing the coupling of remote qubits through shared cavity modes. In this work, we investigate a spin array coupled to a ring cavity supporting two counterpropagating modes, focusing on the system's eigenstates and spin dynamics in the low-excitation regime. We show that, under specific spatial configurations, the spins naturally self-organize into two groups, within which exciton transport is confined. This spin-grouping mechanism preserves coherence between spins across the two groups, and is leveraged to deterministically transfer entangled states between remote spin pairs with additional dynamical addressing. We further propose feasible implementations using atomic qubits or solid-state platforms. Our scheme enables entangling remote spins within a cavity, highlighting potential applications in scalable quantum information processing. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:33:32 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Chang",
""
]
] |
2506.02362 | Xueqi Cheng | Xueqi Cheng, Minxing Zheng, Shixiang Zhu, Yushun Dong | MISLEADER: Defending against Model Extraction with Ensembles of Distilled Models | null | null | null | null | cs.CR cs.AI | http://creativecommons.org/licenses/by/4.0/ | Model extraction attacks aim to replicate the functionality of a black-box model through query access, threatening the intellectual property (IP) of machine-learning-as-a-service (MLaaS) providers. Defending against such attacks is challenging, as it must balance efficiency, robustness, and utility preservation in the real-world scenario. Despite the recent advances, most existing defenses presume that attacker queries have out-of-distribution (OOD) samples, enabling them to detect and disrupt suspicious inputs. However, this assumption is increasingly unreliable, as modern models are trained on diverse datasets and attackers often operate under limited query budgets. As a result, the effectiveness of these defenses is significantly compromised in realistic deployment scenarios. To address this gap, we propose MISLEADER (enseMbles of dIStiLled modEls Against moDel ExtRaction), a novel defense strategy that does not rely on OOD assumptions. MISLEADER formulates model protection as a bilevel optimization problem that simultaneously preserves predictive fidelity on benign inputs and reduces extractability by potential clone models. Our framework combines data augmentation to simulate attacker queries with an ensemble of heterogeneous distilled models to improve robustness and diversity. We further provide a tractable approximation algorithm and derive theoretical error bounds to characterize defense effectiveness. Extensive experiments across various settings validate the utility-preserving and extraction-resistant properties of our proposed defense strategy. Our code is available at https://github.com/LabRAI/MISLEADER. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:37:09 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cheng",
"Xueqi",
""
],
[
"Zheng",
"Minxing",
""
],
[
"Zhu",
"Shixiang",
""
],
[
"Dong",
"Yushun",
""
]
] |
2506.02363 | Tongyu Li | Tongyu Li, Fang Yao | Function-on-function Differential Regression | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Function-on-function regression has been a topic of substantial interest due to its broad applicability, where the relation between functional predictor and response is concerned. In this article, we propose a new framework for modeling the regression mapping that extends beyond integral type, motivated by the prevalence of physical phenomena governed by differential relations, which is referred to as function-on-function differential regression. However, a key challenge lies in representing the differential regression operator, unlike functions that can be expressed by expansions. As a main contribution, we introduce a new notion of model identification involving differential operators, defined through their action on functions. Based on this action-aware identification, we are able to develop a regularization method for estimation using operator reproducing kernel Hilbert spaces. Then a goodness-of-fit test is constructed, which facilitates model checking for differential regression relations. We establish a Bahadur representation for the regression estimator with various theoretical implications, such as the minimax optimality of the proposed estimator, and the validity and consistency of the proposed test. To illustrate the effectiveness of our method, we conduct simulation studies and an application to a real data example on the thermodynamic energy equation. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 01:47:52 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Tongyu",
""
],
[
"Yao",
"Fang",
""
]
] |
2506.02364 | Liang Li | Liang Li, Jianli Zhao, Sheng Fang, Siyu Chen and Hui Sun | A TRPCA-Inspired Deep Unfolding Network for Hyperspectral Image Denoising via Thresholded t-SVD and Top-K Sparse Transformer | 11 pages,6 figures | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Hyperspectral images (HSIs) are often degraded by complex mixed noise during acquisition and transmission, making effective denoising essential for subsequent analysis. Recent hybrid approaches that bridge model-driven and data-driven paradigms have shown great promise. However, most of these approaches lack effective alternation between different priors or modules, resulting in loosely coupled regularization and insufficient exploitation of their complementary strengths. Inspired by tensor robust principal component analysis (TRPCA), we propose a novel deep unfolding network (DU-TRPCA) that enforces stage-wise alternation between two tightly integrated modules: low-rank and sparse. The low-rank module employs thresholded tensor singular value decomposition (t-SVD), providing a widely adopted convex surrogate for tensor low-rankness and has been demonstrated to effectively capture the global spatial-spectral structure of HSIs. The Top-K sparse transformer module adaptively imposes sparse constraints, directly matching the sparse regularization in TRPCA and enabling effective removal of localized outliers and complex noise. This tightly coupled architecture preserves the stage-wise alternation between low-rank approximation and sparse refinement inherent in TRPCA, while enhancing representational capacity through attention mechanisms. Extensive experiments on synthetic and real-world HSIs demonstrate that DU-TRPCA surpasses state-of-the-art methods under severe mixed noise, while offering interpretability benefits and stable denoising dynamics inspired by iterative optimization. Code is available at https://github.com/liangli97/TRPCA-Deep-Unfolding-HSI-Denoising. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:01:39 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Liang",
""
],
[
"Zhao",
"Jianli",
""
],
[
"Fang",
"Sheng",
""
],
[
"Chen",
"Siyu",
""
],
[
"Sun",
"Hui",
""
]
] |
2506.02365 | Chenglou Liu | Chenglou Liu, Yufeng Lu, Fangfang Xie, Tingwei Ji, Yao Zheng | Dynamic real-time multi-UAV cooperative mission planning method under multiple constraints | null | null | null | null | cs.RO cs.SY eess.SY | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | As UAV popularity soars, so does the mission planning associated with it. The classical approaches suffer from the triple problems of decoupled of task assignment and path planning, poor real-time performance and limited adaptability. Aiming at these challenges, this paper proposes a dynamic real-time multi-UAV collaborative mission planning algorithm based on Dubins paths under a distributed formation structure. Dubins path with multiple advantages bridges the gap between task assignment and path planning, leading to a coupled solution for mission planning. Then, a series of acceleration techniques, task clustering preprocessing, highly efficient distance cost functions, low-complexity and less iterative task allocation strategies, are employed to guarantee the real-time performance of the algorithms. To cope with different emergencies and their simultaneous extremes, real-time planning of emerging tasks and mission replanning due to the reduction of available UAVs are appropriately handled. Finally, the developed algorithm is comprehensively exemplified and studied through simulations, highlighting that the proposed method only sacrifices 9.57% of the path length, while achieving a speed improvement of 4-5 orders of magnitude over the simulated annealing method, with a single mission planning of about 0.0003s. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:02:08 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Chenglou",
""
],
[
"Lu",
"Yufeng",
""
],
[
"Xie",
"Fangfang",
""
],
[
"Ji",
"Tingwei",
""
],
[
"Zheng",
"Yao",
""
]
] |
2506.02366 | Qin Xie | Qin Xie, Qinghua Zhang, and Shuyin Xia | Approximate Borderline Sampling using Granular-Ball for Classification Tasks | null | null | null | null | cs.CV cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Data sampling enhances classifier efficiency and robustness through data compression and quality improvement. Recently, the sampling method based on granular-ball (GB) has shown promising performance in generality and noisy classification tasks. However, some limitations remain, including the absence of borderline sampling strategies and issues with class boundary blurring or shrinking due to overlap between GBs. In this paper, an approximate borderline sampling method using GBs is proposed for classification tasks. First, a restricted diffusion-based GB generation (RD-GBG) method is proposed, which prevents GB overlaps by constrained expansion, preserving precise geometric representation of GBs via redefined ones. Second, based on the concept of heterogeneous nearest neighbor, a GB-based approximate borderline sampling (GBABS) method is proposed, which is the first general sampling method capable of both borderline sampling and improving the quality of class noise datasets. Additionally, since RD-GBG incorporates noise detection and GBABS focuses on borderline samples, GBABS performs outstandingly on class noise datasets without the need for an optimal purity threshold. Experimental results demonstrate that the proposed methods outperform the GB-based sampling method and several representative sampling methods. Our source code is publicly available at https://github.com/CherylTse/GBABS. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:04:27 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xie",
"Qin",
""
],
[
"Zhang",
"Qinghua",
""
],
[
"Xia",
"Shuyin",
""
]
] |
2506.02367 | Dequan Jin | Jiayi Su, Dequan Jin | ViTNF: Leveraging Neural Fields to Boost Vision Transformers in Generalized Category Discovery | 22 pages, 3 figures | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Generalized category discovery (GCD) is a highly popular task in open-world recognition, aiming to identify unknown class samples using known class data. By leveraging pre-training, meta-training, and fine-tuning, ViT achieves excellent few-shot learning capabilities. Its MLP head is a feedforward network, trained synchronously with the entire network in the same process, increasing the training cost and difficulty without fully leveraging the power of the feature extractor. This paper proposes a new architecture by replacing the MLP head with a neural field-based one. We first present a new static neural field function to describe the activity distribution of the neural field and then use two static neural field functions to build an efficient few-shot classifier. This neural field-based (NF) classifier consists of two coupled static neural fields. It stores the feature information of support samples by its elementary field, the known categories by its high-level field, and the category information of support samples by its cross-field connections. We replace the MLP head with the proposed NF classifier, resulting in a novel architecture ViTNF, and simplify the three-stage training mode by pre-training the feature extractor on source tasks and training the NF classifier with support samples in meta-testing separately, significantly reducing ViT's demand for training samples and the difficulty of model training. To enhance the model's capability in identifying new categories, we provide an effective algorithm to determine the lateral interaction scale of the elementary field. Experimental results demonstrate that our model surpasses existing state-of-the-art methods on CIFAR-100, ImageNet-100, CUB-200, and Standard Cars, achieving dramatic accuracy improvements of 19\% and 16\% in new and all classes, respectively, indicating a notable advantage in GCD. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:06:01 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Su",
"Jiayi",
""
],
[
"Jin",
"Dequan",
""
]
] |
2506.02368 | Juntao You | Xiaoyan Zhao, Juntao You, Yang Zhang, Wenjie Wang, Hong Cheng, Fuli Feng, See-Kiong Ng, Tat-Seng Chua | NextQuill: Causal Preference Modeling for Enhancing LLM Personalization | null | null | null | null | cs.IR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Personalizing large language models (LLMs) for individual users has become increasingly important as they are progressively integrated into real-world applications to support users' daily lives. However, existing personalization approaches often fail to distinguish which components of model predictions and training data truly reflect user preferences, leading to superficial personalization alignment. In this paper, we introduce NextQuill, a novel LLM personalization alignment framework grounded in causal preference modeling. We approach personalization from a causal perspective, treating both model predictions and ground-truth data generation as outcomes influenced by user preferences, along with other factors. We define the true preference effect as the causal impact of user history (which reflects preferences) on each token prediction or data generation instance, estimated through causal intervention techniques. Building on this insight, NextQuill introduces two complementary alignment strategies: (1) aligning model-internal causal preference effects on predictions with those reflected in ground-truth data, rather than indiscriminately fitting predictions, and (2) focusing on fitting preference-bearing tokens identified via ground-truth data preference effects, rather than treating all tokens uniformly. By integrating these strategies, NextQuill shifts the alignment process toward learning from causal preference effects, facilitating more effective and personalized adaptation. Experiments across multiple personalization benchmarks demonstrate that NextQuill significantly improves personalization quality, offering a principled, causal foundation for LLM personalization. Our codes are available on https://github.com/juntaoyou/NextQuill. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:08:55 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhao",
"Xiaoyan",
""
],
[
"You",
"Juntao",
""
],
[
"Zhang",
"Yang",
""
],
[
"Wang",
"Wenjie",
""
],
[
"Cheng",
"Hong",
""
],
[
"Feng",
"Fuli",
""
],
[
"Ng",
"See-Kiong",
""
],
[
"Chua",
"Tat-Seng",
""
]
] |
2506.02369 | Yuka Kotorii | Senja Barthel, Yuka Kotorii | Linking number of grid models | 14 pages, 1 figure | null | null | null | math.GT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper studies the linking numbers of random links within the grid model. The linking number is treated as a random variable on the isotopy classes of 2-component links, with the paper exploring its asymptotic growth as the diagram size increases. The main result is that the $u$th moment of the linking number for a random link is a polynomial in the grid size with degree $d\leq u$, and all odd moments vanishing. The limits of the moments of the normalized linking number are computed, and it is shown that the distribution of the normalized linking number converges weakly as the grid size tends to infinity. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:12:29 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Barthel",
"Senja",
""
],
[
"Kotorii",
"Yuka",
""
]
] |
2506.02370 | Zhe Li | Zhe Li, Bicheng Ying, Zidong Liu, Chaosheng Dong, Haibo Yang | Reconciling Hessian-Informed Acceleration and Scalar-Only Communication for Efficient Federated Zeroth-Order Fine-Tuning | Under review | null | null | null | cs.LG cs.DC | http://creativecommons.org/licenses/by/4.0/ | Recent dimension-free communication frameworks in Federated Learning (FL), such as DeComFL, significantly reduce per-round communication by transmitting only scalars via zeroth-order stochastic gradient descent (ZO-SGD). This method is particularly advantageous for federated fine-tuning of Large Language Models (LLMs). Yet, the high variance in ZO gradient estimation typically leads to slow convergence. Although leveraging Hessian information is known to enhance optimization speed, integrating this into FL presents significant challenges. These include clients' restrictions on local data and the critical need to maintain the dimension-free communication property. To overcome this limitation, we first introduce a generalized scalar-only communication FL framework that decouples dimension-free communication from standard ZO-SGD, enabling the integration of more advanced optimization strategies. Building on this framework, we propose HiSo, a fast federated fine-tuning method via Hessian-informed zeroth-order optimization and Scalar-only communication. Specifically, it leverages global curvature information to accelerate convergence while preserving the same minimal communication cost per round. Theoretically, we establish convergence guarantees that are independent of the global Lipschitz constant, and further show that HiSo achieves faster rates when the global Hessian exhibits a low effective rank -- a common phenomenon in LLMs. Extensive experiments on benchmark datasets and LLM fine-tuning tasks confirm that HiSo significantly outperforms existing ZO-based FL methods in both convergence speed and communication efficiency. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:13:31 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Zhe",
""
],
[
"Ying",
"Bicheng",
""
],
[
"Liu",
"Zidong",
""
],
[
"Dong",
"Chaosheng",
""
],
[
"Yang",
"Haibo",
""
]
] |
2506.02371 | Haoye Lu Mr. | Haoye Lu, Darren Lo, Yaoliang Yu | SFBD Flow: A Continuous-Optimization Framework for Training Diffusion Models with Noisy Samples | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Diffusion models achieve strong generative performance but often rely on large datasets that may include sensitive content. This challenge is compounded by the models' tendency to memorize training data, raising privacy concerns. SFBD (Lu et al., 2025) addresses this by training on corrupted data and using limited clean samples to capture local structure and improve convergence. However, its iterative denoising and fine-tuning loop requires manual coordination, making it burdensome to implement. We reinterpret SFBD as an alternating projection algorithm and introduce a continuous variant, SFBD flow, that removes the need for alternating steps. We further show its connection to consistency constraint-based methods, and demonstrate that its practical instantiation, Online SFBD, consistently outperforms strong baselines across benchmarks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:16:02 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lu",
"Haoye",
""
],
[
"Lo",
"Darren",
""
],
[
"Yu",
"Yaoliang",
""
]
] |
2506.02372 | Hisami Suzuki | Hisami Suzuki, Satoru Katsumata, Takashi Kodama, Tetsuro Takahashi, Kouta Nakayama, Satoshi Sekine | AnswerCarefully: A Dataset for Improving the Safety of Japanese LLM Output | null | null | null | null | cs.CL | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In this paper we present AnswerCarefully, a dataset for promoting the safety and appropriateness of Japanese LLM outputs. The dataset consists of 1,800 pairs of questions and reference answers, where the questions require special attention in answering. It covers a wide range of risk categories established in prior English-language datasets, but the data samples are original in that they are manually created to reflect the socio-cultural context of LLM usage in Japan. We show that using this dataset for instruction to fine-tune a Japanese LLM led to improved output safety without compromising the utility of general responses. We also report the results of a safety evaluation of 12 Japanese LLMs using this dataset as a benchmark. Finally, we describe the latest update on the dataset which provides English translations and annotations of the questions, aimed at facilitating the derivation of similar datasets in different languages and regions. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:18:59 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Suzuki",
"Hisami",
""
],
[
"Katsumata",
"Satoru",
""
],
[
"Kodama",
"Takashi",
""
],
[
"Takahashi",
"Tetsuro",
""
],
[
"Nakayama",
"Kouta",
""
],
[
"Sekine",
"Satoshi",
""
]
] |
2506.02373 | Kordel France | Kordel K. France, Ovidiu Daescu | Olfactory Inertial Odometry: Methodology for Effective Robot Navigation by Scent | null | null | null | null | cs.RO cs.LG cs.SY eess.SY physics.ins-det | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Olfactory navigation is one of the most primitive mechanisms of exploration used by organisms. Navigation by machine olfaction (artificial smell) is a very difficult task to both simulate and solve. With this work, we define olfactory inertial odometry (OIO), a framework for using inertial kinematics, and fast-sampling olfaction sensors to enable navigation by scent analogous to visual inertial odometry (VIO). We establish how principles from SLAM and VIO can be extrapolated to olfaction to enable real-world robotic tasks. We demonstrate OIO with three different odour localization algorithms on a real 5-DoF robot arm over an odour-tracking scenario that resembles real applications in agriculture and food quality control. Our results indicate success in establishing a baseline framework for OIO from which other research in olfactory navigation can build, and we note performance enhancements that can be made to address more complex tasks in the future. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:21:12 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"France",
"Kordel K.",
""
],
[
"Daescu",
"Ovidiu",
""
]
] |
2506.02374 | Jos\'e Laranjeira | Jose A. S. Laranjeira, Nicolas F. Martins, Kleuton A. L. Lima, Lingtao Xiao, Xihao Chen, Luiz A. Ribeiro Junior, and Julio R. Sambrano | Sodium-Decorated P-C3N: A Porous 2D Framework for High-Capacity and Reversible Hydrogen Storage | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The development of reversible hydrogen storage materials has become crucial for enabling carbon-neutral energy systems. Based on this, the present work investigates the hydrogen storage on the sodium-decorated P-C$_3$N (Na@P-C$_3$N), a porous carbon nitride monolayer recently proposed as a stable semiconductor. First-principles calculations reveal that Na atoms preferentially adsorb with an adsorption energy of -4.48~eV, effectively suppressing clusterization effects. Upon decoration, the system becomes metallic, while \textit{ab initio} molecular dynamics simulations confirm the thermal stability of Na@P-C$_3$N at 300~K. Hydrogen adsorption on Na@P-C$_3$N occurs through weak physisorption, with energies ranging from -0.18 to -0.28~eV, and desorption temperatures between 231 and 357~K. The system can stably absorb 16 H$_2$ molecules per unit cell, corresponding to a gravimetric storage capacity of 9.88~wt\%, surpassing the U.S. Department of Energy target. These results demonstrate that Na@P-C$_3$N is a promising candidate for lightweight, stable, and reversible hydrogen storage. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:21:25 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Laranjeira",
"Jose A. S.",
""
],
[
"Martins",
"Nicolas F.",
""
],
[
"Lima",
"Kleuton A. L.",
""
],
[
"Xiao",
"Lingtao",
""
],
[
"Chen",
"Xihao",
""
],
[
"Junior",
"Luiz A. Ribeiro",
""
],
[
"Sambrano",
"Julio R.",
""
]
] |
2506.02375 | Jos\'e Laranjeira | Jose A. S. Laranjeira, Warda Elaggoune, Nicolas F. Martins, Xihao Chen, and Julio R. Sambrano | OLi3-decorated Irida-graphene for High-capacity Hydrogen Storage: A First-principles Study | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Efficient hydrogen storage in solid-state materials is essential for next-generation energy systems, yet achieving a high gravimetric capacity with optimal adsorption characteristics remains a critical challenge. Although Li-decorated irida-graphene (IG) has shown promising hydrogen storage potential, its capacity is limited to $\sim$ 7wt\%, which, despite exceeding the U.S. DOE target, remains inadequate for large-scale applications. Additionally, Li clustering over extended cycles may compromise adsorption efficiency and structural stability. In this study, we employ first-principles calculations to investigate the hydrogen storage potential of IG decorated with superalkali OLi$_3$ clusters, aiming to enhance the adsorption capacity and stability for advanced hydrogen storage technologies. Our findings show that the OLi$_3$ clusters exhibit a significant binding energy of -3.24 eV, which highlights its strong interaction with the IG. OLi$_3$@IG complex can host up to 12H$_2$ molecules, with optimal maximum storage capacity of 10.00 wt\%. Additionally, the release temperature (T$_R$) and \textit{ab initio} molecular dynamics (AIMD) simulations indicate that H$_2$ molecules can be efficiently released at operating temperatures under ambient conditions. These results highlight the potential of OLi$_3$-decorated irida-graphene as a promising candidate for reversible hydrogen storage. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:21:33 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Laranjeira",
"Jose A. S.",
""
],
[
"Elaggoune",
"Warda",
""
],
[
"Martins",
"Nicolas F.",
""
],
[
"Chen",
"Xihao",
""
],
[
"Sambrano",
"Julio R.",
""
]
] |
2506.02376 | Xin Yong | Didarul Ahasan Redwan and Justin Reicher and Xin Yong | Cell-Scale Dynamic Modeling of Membrane Interactions with Arbitrarily Shaped Particles | 21 pages, 8 figures | null | null | null | cond-mat.soft physics.bio-ph | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Modeling membrane interactions with arbitrarily shaped colloidal particles, such as environmental micro- and nanoplastics, at the cell scale remains particularly challenging, owing to the complexity of particle geometries and the need to resolve fully coupled translational and rotational dynamics. Here, we present a force-based computational framework capable of capturing dynamic interactions between deformable lipid vesicles and rigid particles of irregular shapes. Both vesicle and particle surfaces are represented using triangulated meshes, and Langevin dynamics resolves membrane deformation alongside rigid-body particle motion. Adhesive interactions between the particle and membrane surfaces are modeled using two numerical schemes: a vertex-to-vertex mapping and a vertex-to-surface projection. The latter yields more accurate wrapping energetics, as demonstrated by benchmark comparisons against ideal spheres. The dynamic simulations reveal that lower particle-to-vesicle mass ratios facilitate frequent particle reorientation and complete membrane wrapping, while higher mass ratios limit orientation changes and stabilize partial wrapping. To illustrate the framework's versatility, we simulate interactions involving cubical, rod-like, bowl-shaped, and tetrahedral particles with spherical, cigar-shaped, or biconcave vesicles. This generalizable modeling approach enables predictive, cell-scale studies of membrane-particle interactions across a wide range of geometries, with applications in environmental biophysics and nanomedicine. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:22:21 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Redwan",
"Didarul Ahasan",
""
],
[
"Reicher",
"Justin",
""
],
[
"Yong",
"Xin",
""
]
] |
2506.02377 | Zain Mehdi Dr | Zain Mehdi and Matthew L. Goh and Matthew J. Blacker and Joseph J. Hope and Stuart S. Szigeti | Multi-mode cooling of a Bose-Einstein condensate with linear quantum feedback | null | null | null | null | cond-mat.quant-gas quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We theoretically investigate measurement-based feedback control over the motional degrees of freedom of an oblate quasi-2D atomic Bose-Einstein condensate (BEC) subject to continuous density monitoring. We develop a linear-quadratic-Gaussian (LQG) model that describes the multi-mode dynamics of the condensate's collective excitations under continuous measurement and control. Crucially, the multi-mode cold-damping feedback control we consider uses a realistic state-estimation scheme that does not rely upon a particular model of the atomic dynamics. We present analytical results showing that collective excitations can be cooled to below single-phonon average occupation (ground-state cooling) across a broad parameter regime and identify the conditions under which the lowest steady-state phonon occupation is asymptotically achieved. Further, we develop multi-objective optimization methods that explore the trade-off between cooling speed and the final energy of the cloud and provide numerical simulations demonstrating the ground-state cooling of the lowest ten motional modes above the condensate ground state. Our investigation provides concrete guidance on the feedback control design and parameters needed to experimentally realize a feedback-cooled BEC. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:26:48 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Mehdi",
"Zain",
""
],
[
"Goh",
"Matthew L.",
""
],
[
"Blacker",
"Matthew J.",
""
],
[
"Hope",
"Joseph J.",
""
],
[
"Szigeti",
"Stuart S.",
""
]
] |
2506.02378 | Ukyo Honda | Ukyo Honda and Tatsushi Oka | Exploring Explanations Improves the Robustness of In-Context Learning | Accepted to ACL 2025 (Main Conference) | null | null | null | cs.CL cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | In-context learning (ICL) has emerged as a successful paradigm for leveraging large language models (LLMs). However, it often struggles to generalize beyond the distribution of the provided demonstrations. A recent advancement in enhancing robustness is ICL with explanations (X-ICL), which improves prediction reliability by guiding LLMs to understand and articulate the reasoning behind correct labels. Building on this approach, we introduce an advanced framework that extends X-ICL by systematically exploring explanations for all possible labels (X$^2$-ICL), thereby enabling more comprehensive and robust decision-making. Experimental results on multiple natural language understanding datasets validate the effectiveness of X$^2$-ICL, demonstrating significantly improved robustness to out-of-distribution data compared to the existing ICL approaches. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:29:14 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Honda",
"Ukyo",
""
],
[
"Oka",
"Tatsushi",
""
]
] |
2506.02379 | Hideya Watanabe | Hideya Watanabe | Finite-dimensional irreducible representations of twisted loop algebras of the second kind | 36 pages | null | null | null | math.RT | http://creativecommons.org/licenses/by/4.0/ | Twisted loop algebras of the second kind are infinite-dimensional Lie algebras that are constructed from a semisimple Lie algebra and an automorphism on it of order at most $2$. They are examples of equivariant map algebras. The finite-dimensional irreducible representations of an arbitrary equivariant map algebra have been classified by Neher--Savage--Senesi. In this paper, we classify the finite-dimensional irreducible representations of twisted loop algebras of the second kind in a more elementary way. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:32:07 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Watanabe",
"Hideya",
""
]
] |
2506.02380 | Zihao Ding | Zihao Ding, Cheng-Tse Lee, Mufeng Zhu, Tao Guan, Yuan-Chun Sun, Cheng-Hsin Hsu, Yao Liu | EyeNavGS: A 6-DoF Navigation Dataset and Record-n-Replay Software for Real-World 3DGS Scenes in VR | null | null | null | null | cs.MM cs.CV cs.GR cs.HC | http://creativecommons.org/licenses/by/4.0/ | 3D Gaussian Splatting (3DGS) is an emerging media representation that reconstructs real-world 3D scenes in high fidelity, enabling 6-degrees-of-freedom (6-DoF) navigation in virtual reality (VR). However, developing and evaluating 3DGS-enabled applications and optimizing their rendering performance, require realistic user navigation data. Such data is currently unavailable for photorealistic 3DGS reconstructions of real-world scenes. This paper introduces EyeNavGS (EyeNavGS), the first publicly available 6-DoF navigation dataset featuring traces from 46 participants exploring twelve diverse, real-world 3DGS scenes. The dataset was collected at two sites, using the Meta Quest Pro headsets, recording the head pose and eye gaze data for each rendered frame during free world standing 6-DoF navigation. For each of the twelve scenes, we performed careful scene initialization to correct for scene tilt and scale, ensuring a perceptually-comfortable VR experience. We also release our open-source SIBR viewer software fork with record-and-replay functionalities and a suite of utility tools for data processing, conversion, and visualization. The EyeNavGS dataset and its accompanying software tools provide valuable resources for advancing research in 6-DoF viewport prediction, adaptive streaming, 3D saliency, and foveated rendering for 3DGS scenes. The EyeNavGS dataset is available at: https://symmru.github.io/EyeNavGS/. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:32:35 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ding",
"Zihao",
""
],
[
"Lee",
"Cheng-Tse",
""
],
[
"Zhu",
"Mufeng",
""
],
[
"Guan",
"Tao",
""
],
[
"Sun",
"Yuan-Chun",
""
],
[
"Hsu",
"Cheng-Hsin",
""
],
[
"Liu",
"Yao",
""
]
] |
2506.02381 | Fei Chen | Songlin Wei, Gene Cheung, Fei Chen, Ivan Selesnick | Unrolling Nonconvex Graph Total Variation for Image Denoising | null | null | null | null | eess.IV cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Conventional model-based image denoising optimizations employ convex regularization terms, such as total variation (TV) that convexifies the $\ell_0$-norm to promote sparse signal representation. Instead, we propose a new non-convex total variation term in a graph setting (NC-GTV), such that when combined with an $\ell_2$-norm fidelity term for denoising, leads to a convex objective with no extraneous local minima. We define NC-GTV using a new graph variant of the Huber function, interpretable as a Moreau envelope. The crux is the selection of a parameter $a$ characterizing the graph Huber function that ensures overall objective convexity; we efficiently compute $a$ via an adaptation of Gershgorin Circle Theorem (GCT). To minimize the convex objective, we design a linear-time algorithm based on Alternating Direction Method of Multipliers (ADMM) and unroll it into a lightweight feed-forward network for data-driven parameter learning. Experiments show that our method outperforms unrolled GTV and other representative image denoising schemes, while employing far fewer network parameters. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:34:32 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wei",
"Songlin",
""
],
[
"Cheung",
"Gene",
""
],
[
"Chen",
"Fei",
""
],
[
"Selesnick",
"Ivan",
""
]
] |
2506.02382 | Seulgi Kim | Seulgi Kim, Ghazal Kaviani, Mohit Prabhushankar, Ghassan AlRegib | Multi-level and Multi-modal Action Anticipation | Accepted in 2025 IEEE International Conference on Image Processing (ICIP) | null | null | null | cs.CV cs.LG | http://creativecommons.org/licenses/by/4.0/ | Action anticipation, the task of predicting future actions from partially observed videos, is crucial for advancing intelligent systems. Unlike action recognition, which operates on fully observed videos, action anticipation must handle incomplete information. Hence, it requires temporal reasoning, and inherent uncertainty handling. While recent advances have been made, traditional methods often focus solely on visual modalities, neglecting the potential of integrating multiple sources of information. Drawing inspiration from human behavior, we introduce \textit{Multi-level and Multi-modal Action Anticipation (m\&m-Ant)}, a novel multi-modal action anticipation approach that combines both visual and textual cues, while explicitly modeling hierarchical semantic information for more accurate predictions. To address the challenge of inaccurate coarse action labels, we propose a fine-grained label generator paired with a specialized temporal consistency loss function to optimize performance. Extensive experiments on widely used datasets, including Breakfast, 50 Salads, and DARai, demonstrate the effectiveness of our approach, achieving state-of-the-art results with an average anticipation accuracy improvement of 3.08\% over existing methods. This work underscores the potential of multi-modal and hierarchical modeling in advancing action anticipation and establishes a new benchmark for future research in the field. Our code is available at: https://github.com/olivesgatech/mM-ant. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:39:33 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kim",
"Seulgi",
""
],
[
"Kaviani",
"Ghazal",
""
],
[
"Prabhushankar",
"Mohit",
""
],
[
"AlRegib",
"Ghassan",
""
]
] |
2506.02383 | Carlos Arnoldo Morales Rojas | E. Rego, C. Rojas, X. Wen | Rescaled topological entropy | 18 pages, 2 figures | null | null | null | math.DS | http://creativecommons.org/licenses/by/4.0/ | We prove that to any smooth vector field of a closed manifold it can be assigned a nonnegative number called {\em rescaled topological entropy} satisfying the following properties: it is an upper bound for both the topological entropy and the rescaled metric entropy \cite{ww}; coincides with the topological entropy for nonsingular vector fields; is positive for certain surface vector fields (in contrast to the topological entropy); is invariant under rescaled topological conjugacy; and serves as an upper bound for the growth rate of periodic orbits for rescaling expansive flows with dynamically isolated singular set. Therefore, the rescaled topological entropy bounds such growth rates for $C^r$-generic rescaling (or $k^*$) expansive vector fields on closed manifolds. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:50:06 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Rego",
"E.",
""
],
[
"Rojas",
"C.",
""
],
[
"Wen",
"X.",
""
]
] |
2506.02384 | Dan M. Kluger | Dan M. Kluger and Stefania Di Tommaso and David B. Lobell | Evaluating crop rotations around the world using satellite imagery and causal machine learning | null | null | null | null | stat.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Building sustainable food systems that are resilient to climate change will require improved agricultural management and policy. One common practice that is well-known to benefit crop yields is crop rotation, yet there remains limited understanding of how the benefits of crop rotation vary for different crop sequences and under different weather conditions. To address these gaps, we leverage crop type maps, satellite data, and causal machine learning to study how precrop effects on subsequent yields vary with cropping sequence choice and weather. Complementing and going beyond what is known from randomized field trials, we find that (i) for those farmers who do rotate, the most common precrop choices tend to be among the most beneficial, (ii) the effects of switching from a simple rotation (which alternates between two crops) to a more diverse rotation were typically small and sometimes even negative, (iii) precrop effects tended to be greater under rainier conditions, (iv) precrop effects were greater under warmer conditions for soybean yields but not for other crops, and (v) legume precrops conferred smaller benefits under warmer conditions. Our results and the methods we use can enable farmers and policy makers to identify which rotations will be most effective at improving crop yields in a changing climate. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:51:56 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kluger",
"Dan M.",
""
],
[
"Di Tommaso",
"Stefania",
""
],
[
"Lobell",
"David B.",
""
]
] |
2506.02385 | Shuze Chen | Shuze Chen, Tianyi Peng | Multi-agent Markov Entanglement | null | null | null | null | cs.LG stat.ML | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Value decomposition has long been a fundamental technique in multi-agent dynamic programming and reinforcement learning (RL). Specifically, the value function of a global state $(s_1,s_2,\ldots,s_N)$ is often approximated as the sum of local functions: $V(s_1,s_2,\ldots,s_N)\approx\sum_{i=1}^N V_i(s_i)$. This approach traces back to the index policy in restless multi-armed bandit problems and has found various applications in modern RL systems. However, the theoretical justification for why this decomposition works so effectively remains underexplored.
In this paper, we uncover the underlying mathematical structure that enables value decomposition. We demonstrate that a multi-agent Markov decision process (MDP) permits value decomposition if and only if its transition matrix is not "entangled" -- a concept analogous to quantum entanglement in quantum physics. Drawing inspiration from how physicists measure quantum entanglement, we introduce how to measure the "Markov entanglement" for multi-agent MDPs and show that this measure can be used to bound the decomposition error in general multi-agent MDPs.
Using the concept of Markov entanglement, we proved that a widely-used class of index policies is weakly entangled and enjoys a sublinear $\mathcal O(\sqrt{N})$ scale of decomposition error for $N$-agent systems. Finally, we show how Markov entanglement can be efficiently estimated in practice, providing practitioners with an empirical proxy for the quality of value decomposition. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:54:25 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Shuze",
""
],
[
"Peng",
"Tianyi",
""
]
] |
2506.02386 | Vincent Tan | Jie Bian and Vincent Y. F. Tan | Asymptotically Optimal Linear Best Feasible Arm Identification with Fixed Budget | Accepted to the Conference on Uncertainty in Artificial Intelligence (UAI) 2025 | null | null | null | cs.LG cs.AI cs.IT math.IT | http://creativecommons.org/licenses/by/4.0/ | The challenge of identifying the best feasible arm within a fixed budget has attracted considerable interest in recent years. However, a notable gap remains in the literature: the exact exponential rate at which the error probability approaches zero has yet to be established, even in the relatively simple setting of $K$-armed bandits with Gaussian noise. In this paper, we address this gap by examining the problem within the context of linear bandits. We introduce a novel algorithm for best feasible arm identification that guarantees an exponential decay in the error probability. Remarkably, the decay rate -- characterized by the exponent -- matches the theoretical lower bound derived using information-theoretic principles. Our approach leverages a posterior sampling framework embedded within a game-based sampling rule involving a min-learner and a max-learner. This strategy shares its foundations with Thompson sampling, but is specifically tailored to optimize the identification process under fixed-budget constraints. Furthermore, we validate the effectiveness of our algorithm through comprehensive empirical evaluations across various problem instances with different levels of complexity. The results corroborate our theoretical findings and demonstrate that our method outperforms several benchmark algorithms in terms of both accuracy and efficiency. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:56:26 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bian",
"Jie",
""
],
[
"Tan",
"Vincent Y. F.",
""
]
] |
2506.02387 | Zelai Xu | Zelai Xu, Zhexuan Xu, Xiangmin Yi, Huining Yuan, Xinlei Chen, Yi Wu, Chao Yu, Yu Wang | VS-Bench: Evaluating VLMs for Strategic Reasoning and Decision-Making in Multi-Agent Environments | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by/4.0/ | Recent advancements in Vision Language Models (VLMs) have expanded their capabilities to interactive agent tasks, yet existing benchmarks remain limited to single-agent or text-only environments. In contrast, real-world scenarios often involve multiple agents interacting within rich visual and linguistic contexts, posing challenges with both multimodal observations and strategic interactions. To bridge this gap, we introduce Visual Strategic Bench (VS-Bench), a multimodal benchmark that evaluates VLMs for strategic reasoning and decision-making in multi-agent environments. VS-Bench comprises eight vision-grounded environments spanning cooperative, competitive, and mixed-motive interactions, designed to assess agents' ability to predict others' future moves and optimize for long-term objectives. We consider two complementary evaluation dimensions, including offline evaluation of strategic reasoning by next-action prediction accuracy and online evaluation of decision-making by normalized episode return. Extensive experiments of fourteen leading VLMs reveal a significant gap between current models and optimal performance, with the best models attaining 47.8% prediction accuracy and 24.3% normalized return. We further conduct in-depth analyses on multimodal observations, test-time scaling, social behaviors, and failure cases of VLM agents. By standardizing the evaluation and highlighting the limitations of existing models, we envision VS-Bench as a foundation for future research on strategic multimodal agents. Code and data are available at https://vs-bench.github.io. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:57:38 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xu",
"Zelai",
""
],
[
"Xu",
"Zhexuan",
""
],
[
"Yi",
"Xiangmin",
""
],
[
"Yuan",
"Huining",
""
],
[
"Chen",
"Xinlei",
""
],
[
"Wu",
"Yi",
""
],
[
"Yu",
"Chao",
""
],
[
"Wang",
"Yu",
""
]
] |
2506.02388 | Justin Bloom | Justin Bloom | Families of Green rings for abelian restricted Lie algebras | 29 pages | null | null | null | math.RT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We find equivalent conditions determining the representation type of abelian restricted Lie algebras in terms of how their Green ring of restricted representations varies with respect to different cocommutative Hopf algebra structures on its restricted universal enveloping algebra. Each compatible cocommutative Hopf algebra structure on a tame algebra is shown to have a correspondence between a certain set of Hopf subalgebras, and the set of minimal thick tensor-ideals having identical ring structure when determined by either the Hopf algebra structure or the base Lie algebra structure (up to a choice of character group). Those of wild representation type are shown never to have such a correspondence. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 02:58:52 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bloom",
"Justin",
""
]
] |
2506.02389 | Chamara Madarasingha Dr | Chamara Madarasingha, Nasrin Sohrabi, Zahir Tari | Univariate to Multivariate: LLMs as Zero-Shot Predictors for Time-Series Forecasting | null | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by/4.0/ | Time-series prediction or forecasting is critical across many real-world dynamic systems, and recent studies have proposed using Large Language Models (LLMs) for this task due to their strong generalization capabilities and ability to perform well without extensive pre-training. However, their effectiveness in handling complex, noisy, and multivariate time-series data remains underexplored. To address this, we propose LLMPred which enhances LLM-based time-series prediction by converting time-series sequences into text and feeding them to LLMs for zero shot prediction along with two main data pre-processing techniques. First, we apply time-series sequence decomposition to facilitate accurate prediction on complex and noisy univariate sequences. Second, we extend this univariate prediction capability to multivariate data using a lightweight prompt-processing strategy. Extensive experiments with smaller LLMs such as Llama 2 7B, Llama 3.2 3B, GPT-4o-mini, and DeepSeek 7B demonstrate that LLMPred achieves competitive or superior performance compared to state-of-the-art baselines. Additionally, a thorough ablation study highlights the importance of the key components proposed in LLMPred. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:02:47 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Madarasingha",
"Chamara",
""
],
[
"Sohrabi",
"Nasrin",
""
],
[
"Tari",
"Zahir",
""
]
] |
2506.02390 | Qin Xie | Qin Xie, Qinghua Zhang, Shuyin Xia, Xinran Zhou, and Guoyin Wang | GAdaBoost: An Efficient and Robust AdaBoost Algorithm Based on Granular-Ball Structure | null | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Adaptive Boosting (AdaBoost) faces significant challenges posed by label noise, especially in multiclass classification tasks. Existing methods either lack mechanisms to handle label noise effectively or suffer from high computational costs due to redundant data usage. Inspired by granular computing, this paper proposes granular adaptive boosting (GAdaBoost), a novel two-stage framework comprising a data granulation stage and an adaptive boosting stage, to enhance efficiency and robustness under noisy conditions. To validate its feasibility, an extension of SAMME, termed GAdaBoost.SA, is proposed. Specifically, first, a granular-ball generation method is designed to compress data while preserving diversity and mitigating label noise. Second, the granular ball-based SAMME algorithm focuses on granular balls rather than individual samples, improving efficiency and reducing sensitivity to noise. Experimental results on some noisy datasets show that the proposed approach achieves superior robustness and efficiency compared with existing methods, demonstrating that this work effectively extends AdaBoost and SAMME. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:10:20 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xie",
"Qin",
""
],
[
"Zhang",
"Qinghua",
""
],
[
"Xia",
"Shuyin",
""
],
[
"Zhou",
"Xinran",
""
],
[
"Wang",
"Guoyin",
""
]
] |
2506.02391 | Chuanghao Ding | Chuanghao Ding, Jiaping Wang, Ziqing Yang, Xiaoliang Wang, Dahua Lin, Cam-Tu Nguyen, Fei Tan | Consultant Decoding: Yet Another Synergistic Mechanism | ACL 2025 findings | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | The synergistic mechanism based on Speculative Decoding (SD) has garnered considerable attention as a simple yet effective approach for accelerating the inference of large language models (LLMs). Nonetheless, the high rejection rates require repeated LLMs calls to validate draft tokens, undermining the overall efficiency gain of SD. In this work, we revisit existing verification mechanisms and propose a novel synergetic mechanism Consultant Decoding (CD). Unlike SD, which relies on a metric derived from importance sampling for verification, CD verifies candidate drafts using token-level likelihoods computed solely by the LLM. CD achieves up to a 2.5-fold increase in inference speed compared to the target model, while maintaining comparable generation quality (around 100% of the target model's performance). Interestingly, this is achieved by combining models whose parameter sizes differ by two orders of magnitude. In addition, CD reduces the call frequency of the large target model to below 10%, particularly in more demanding tasks. CD's performance was even found to surpass that of the large target model, which theoretically represents the upper bound for speculative decoding. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:13:27 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ding",
"Chuanghao",
""
],
[
"Wang",
"Jiaping",
""
],
[
"Yang",
"Ziqing",
""
],
[
"Wang",
"Xiaoliang",
""
],
[
"Lin",
"Dahua",
""
],
[
"Nguyen",
"Cam-Tu",
""
],
[
"Tan",
"Fei",
""
]
] |
2506.02392 | Yuanyao Chen | Yuanyao Chen, Rongsheng Chen, Fu Luo, Zhenkun Wang | Improving Generalization of Neural Combinatorial Optimization for Vehicle Routing Problems via Test-Time Projection Learning | arXiv admin note: text overlap with arXiv:2505.24627 | null | null | null | cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Neural Combinatorial Optimization (NCO) has emerged as a promising learning-based paradigm for addressing Vehicle Routing Problems (VRPs) by minimizing the need for extensive manual engineering. While existing NCO methods, trained on small-scale instances (e.g., 100 nodes), have demonstrated considerable success on problems of similar scale, their performance significantly degrades when applied to large-scale scenarios. This degradation arises from the distributional shift between training and testing data, rendering policies learned on small instances ineffective for larger problems. To overcome this limitation, we introduce a novel learning framework driven by Large Language Models (LLMs). This framework learns a projection between the training and testing distributions, which is then deployed to enhance the scalability of the NCO model. Notably, unlike prevailing techniques that necessitate joint training with the neural network, our approach operates exclusively during the inference phase, obviating the need for model retraining. Extensive experiments demonstrate that our method enables a backbone model (trained on 100-node instances) to achieve superior performance on large-scale Traveling Salesman Problem (TSP) and Capacitated Vehicle Routing Problem (CVRP) of up to 100K nodes from diverse distributions. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:15:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Yuanyao",
""
],
[
"Chen",
"Rongsheng",
""
],
[
"Luo",
"Fu",
""
],
[
"Wang",
"Zhenkun",
""
]
] |
2506.02393 | Yongxian Liu | Yongxian Liu, Boyang Li, Ting Liu, Zaiping Lin, Wei An | RRCANet: Recurrent Reusable-Convolution Attention Network for Infrared Small Target Detection | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Infrared small target detection is a challenging task due to its unique characteristics (e.g., small, dim, shapeless and changeable). Recently published CNN-based methods have achieved promising performance with heavy feature extraction and fusion modules. To achieve efficient and effective detection, we propose a recurrent reusable-convolution attention network (RRCA-Net) for infrared small target detection. Specifically, RRCA-Net incorporates reusable-convolution block (RuCB) in a recurrent manner without introducing extra parameters. With the help of the repetitive iteration in RuCB, the high-level information of small targets in the deep layers can be well maintained and further refined. Then, a dual interactive attention aggregation module (DIAAM) is proposed to promote the mutual enhancement and fusion of refined information. In this way, RRCA-Net can both achieve high-level feature refinement and enhance the correlation of contextual information between adjacent layers. Moreover, to achieve steady convergence, we design a target characteristic inspired loss function (DpT-k loss) by integrating physical and mathematical constraints. Experimental results on three benchmark datasets (e.g. NUAA-SIRST, IRSTD-1k, DenseSIRST) demonstrate that our RRCA-Net can achieve comparable performance to the state-of-the-art methods while maintaining a small number of parameters, and act as a plug and play module to introduce consistent performance improvement for several popular IRSTD methods. Our code will be available at https://github.com/yongxianLiu/ soon. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:18:17 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Yongxian",
""
],
[
"Li",
"Boyang",
""
],
[
"Liu",
"Ting",
""
],
[
"Lin",
"Zaiping",
""
],
[
"An",
"Wei",
""
]
] |
2506.02394 | Yuan Bian | Yuan Bian, Xingche Guo, Yuanjia Wang | Joint Modeling for Learning Decision-Making Dynamics in Behavioral Experiments | null | null | null | null | stat.ME cs.LG | http://creativecommons.org/licenses/by/4.0/ | Major depressive disorder (MDD), a leading cause of disability and mortality, is associated with reward-processing abnormalities and concentration issues. Motivated by the probabilistic reward task from the Establishing Moderators and Biosignatures of Antidepressant Response in Clinical Care (EMBARC) study, we propose a novel framework that integrates the reinforcement learning (RL) model and drift-diffusion model (DDM) to jointly analyze reward-based decision-making with response times. To account for emerging evidence suggesting that decision-making may alternate between multiple interleaved strategies, we model latent state switching using a hidden Markov model (HMM). In the ''engaged'' state, decisions follow an RL-DDM, simultaneously capturing reward processing, decision dynamics, and temporal structure. In contrast, in the ''lapsed'' state, decision-making is modeled using a simplified DDM, where specific parameters are fixed to approximate random guessing with equal probability. The proposed method is implemented using a computationally efficient generalized expectation-maximization algorithm with forward-backward procedures. Through extensive numerical studies, we demonstrate that our proposed method outperforms competing approaches under various reward-generating distributions, both with and without strategy switching. When applied to the EMBARC study, our framework reveals that MDD patients exhibit lower overall engagement than healthy controls and experience longer decision times when they do engage. Additionally, we show that neuroimaging measures of brain activities are associated with decision-making characteristics in the ''engaged'' state but not in the ''lapsed'' state, providing evidence of brain-behavioral association specific to the ''engaged'' state. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:21:10 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bian",
"Yuan",
""
],
[
"Guo",
"Xingche",
""
],
[
"Wang",
"Yuanjia",
""
]
] |
2506.02395 | Xiaofeng Cong | Xiaofeng Cong, Yu-Xin Zhang, Haoran Wei, Yeying Jin, Junming Hou, Jie Gui, Jing Zhang, Dacheng Tao | The Devil is in the Darkness: Diffusion-Based Nighttime Dehazing Anchored in Brightness Perception | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | While nighttime image dehazing has been extensively studied, converting nighttime hazy images to daytime-equivalent brightness remains largely unaddressed. Existing methods face two critical limitations: (1) datasets overlook the brightness relationship between day and night, resulting in the brightness mapping being inconsistent with the real world during image synthesis; and (2) models do not explicitly incorporate daytime brightness knowledge, limiting their ability to reconstruct realistic lighting. To address these challenges, we introduce the Diffusion-Based Nighttime Dehazing (DiffND) framework, which excels in both data synthesis and lighting reconstruction. Our approach starts with a data synthesis pipeline that simulates severe distortions while enforcing brightness consistency between synthetic and real-world scenes, providing a strong foundation for learning night-to-day brightness mapping. Next, we propose a restoration model that integrates a pre-trained diffusion model guided by a brightness perception network. This design harnesses the diffusion model's generative ability while adapting it to nighttime dehazing through brightness-aware optimization. Experiments validate our dataset's utility and the model's superior performance in joint haze removal and brightness mapping. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:21:13 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cong",
"Xiaofeng",
""
],
[
"Zhang",
"Yu-Xin",
""
],
[
"Wei",
"Haoran",
""
],
[
"Jin",
"Yeying",
""
],
[
"Hou",
"Junming",
""
],
[
"Gui",
"Jie",
""
],
[
"Zhang",
"Jing",
""
],
[
"Tao",
"Dacheng",
""
]
] |
2506.02396 | Longyu Yang | Longyu Yang, Ping Hu, Shangbo Yuan, Lu Zhang, Jun Liu, Hengtao Shen, Xiaofeng Zhu | Towards Explicit Geometry-Reflectance Collaboration for Generalized LiDAR Segmentation in Adverse Weather | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Existing LiDAR semantic segmentation models often suffer from decreased accuracy when exposed to adverse weather conditions. Recent methods addressing this issue focus on enhancing training data through weather simulation or universal augmentation techniques. However, few works have studied the negative impacts caused by the heterogeneous domain shifts in the geometric structure and reflectance intensity of point clouds. In this paper, we delve into this challenge and address it with a novel Geometry-Reflectance Collaboration (GRC) framework that explicitly separates feature extraction for geometry and reflectance. Specifically, GRC employs a dual-branch architecture designed to independently process geometric and reflectance features initially, thereby capitalizing on their distinct characteristic. Then, GRC adopts a robust multi-level feature collaboration module to suppress redundant and unreliable information from both branches. Consequently, without complex simulation or augmentation, our method effectively extracts intrinsic information about the scene while suppressing interference, thus achieving better robustness and generalization in adverse weather conditions. We demonstrate the effectiveness of GRC through comprehensive experiments on challenging benchmarks, showing that our method outperforms previous approaches and establishes new state-of-the-art results. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:23:43 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yang",
"Longyu",
""
],
[
"Hu",
"Ping",
""
],
[
"Yuan",
"Shangbo",
""
],
[
"Zhang",
"Lu",
""
],
[
"Liu",
"Jun",
""
],
[
"Shen",
"Hengtao",
""
],
[
"Zhu",
"Xiaofeng",
""
]
] |
2506.02397 | Shengjia Zhang | Shengjia Zhang, Junjie Wu, Jiawei Chen, Changwang Zhang, Xingyu Lou, Wangchunshu Zhou, Sheng Zhou, Can Wang, Jun Wang | OThink-R1: Intrinsic Fast/Slow Thinking Mode Switching for Over-Reasoning Mitigation | null | null | null | null | cs.AI | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Recent advanced large reasoning models (LRMs) leverage extended chain-of-thought (CoT) reasoning to solve complex tasks, achieving state-of-the-art performance. Despite their success, we identify a critical issue: a substantial portion of simple tasks solved by LRMs can also be addressed by non-reasoning LLMs using significantly fewer tokens, indicating the complex reasoning may not always be necessary. To address this, we systematically analyze the reasoning trajectories of LRMs and present a method utilizing identified paradigms and LLM-Judge to classify these trajectories as either Redundant Reasoning or Essential Reasoning. And we introduce OThink-R1, a method that prunes redundant reasoning steps while preserving logical validity. OThink-R1 dynamically employs the non-thinking mode (fast-thinking) for straightforward problems while engaging in deliberate thinking (slow-thinking) for complex problems. Experiments across mathematical and question-answering tasks demonstrate that OThink-R1 reduces reasoning redundancy by almost 23\% on average without compromising accuracy, offering practical guidelines for efficient reasoning models. The code is available at https://github.com/AgenticIR-Lab/OThink-R1. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:31:30 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhang",
"Shengjia",
""
],
[
"Wu",
"Junjie",
""
],
[
"Chen",
"Jiawei",
""
],
[
"Zhang",
"Changwang",
""
],
[
"Lou",
"Xingyu",
""
],
[
"Zhou",
"Wangchunshu",
""
],
[
"Zhou",
"Sheng",
""
],
[
"Wang",
"Can",
""
],
[
"Wang",
"Jun",
""
]
] |
2506.02398 | Prashanth Sridhar | Michael K. Brown, Andrew J. Soto Levins, Prashanth Sridhar | Existence of balanced dualizing dg-modules | 13 pages | null | null | null | math.RA math.AG | http://creativecommons.org/licenses/by/4.0/ | We describe cohomological conditions that are necessary and sufficient for the existence of balanced dualizing dg-modules, generalizing a theorem of Van den Bergh for balanced dualizing complexes over graded algebras. As a consequence, we show that a dg-algebra satisfying certain finiteness conditions admits a balanced dualizing dg-module if and only if its zeroth cohomology algebra admits a balanced dualizing complex. Additionally, we obtain a host of new examples of dg-algebras whose associated noncommutative spaces satisfy Serre duality. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:32:22 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Brown",
"Michael K.",
""
],
[
"Levins",
"Andrew J. Soto",
""
],
[
"Sridhar",
"Prashanth",
""
]
] |
2506.02399 | Guangjie Li | Jordan Gaines, Guangjie Li, Jukka V\"ayrynen | Spin-chain multichannel Kondo model via image impurity boundary condition | 9 pages, 5 figures | null | null | null | cond-mat.str-el | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | One of the signature observables for the electronic multichannel Kondo model is the impurity entropy, which was claimed to be found in $J_1$-$J_2$ Heisenberg chain with open boundary condition (OBC) and periodic boundary condition (PBC), respectively for the one-channel and two-channel case. However, it is not clear how to generalize OBC and PBC in Heisenberg chains to find the multichannel Kondo impurity entropy with more than two channels. In this paper, we demonstrate that the correct boundary condition for realizing multichannel Kondo physics in Heisenberg chains is the image impurity boundary condition (IIBC), which yields the expected impurity entropy, $\ln[(\sqrt{5}+1)/2]$ for the three-channel case and $\ln\sqrt{3}$ for the four-channel case. Moreover, the IIBC reduces to OBC for the one-channel case and to PBC for the two-channel case. With IIBC, the finite-size scaling of the impurity entropy and the total spin both match the finite-temperature correction in the electronic multichannel Kondo model. Additionally, we show that the XXZ anisotropy reduces the impurity entropy through a power law of the effective Luttinger liquid parameter. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:36:03 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Gaines",
"Jordan",
""
],
[
"Li",
"Guangjie",
""
],
[
"Väyrynen",
"Jukka",
""
]
] |
2506.02400 | Andrew Eberhardt | Andrew Eberhardt, Lam Hui | Caustic fringes for wave dark matter | null | null | null | null | astro-ph.CO hep-ph | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Wave dark matter is composed of particles sufficiently light that their de Broglie wavelength exceeds the average inter-particle separation. A typical wave dark matter halo exhibits granular substructures due to wave interference. In this paper, we explore the wave interference effects around caustics. These are locations of formally divergent density in cold collisionless systems. Examples include splashback in galaxy clusters, and tidal shells in merging galaxies, where the pile-up of dark matter close to apogee gives rise to caustics. We show that wave interference modifies the density profile in the vicinity of the caustics, giving rise to a fringe pattern well-described by the Airy function. This follows from approximating the gravitational potential as linear close to apogee. This prediction is verified in a series of numerical simulations in which the gravitational potential is computed exactly. We provide a formula expressing the fringe separation in terms of the wave dark matter mass and halo parameters, which is useful for interpreting and stacking data. The fringe separation near caustics can be significantly larger than the naive de Broglie scale (the latter set by the system's velocity dispersion). This opens up the possibility of detecting caustic fringes for a wide range of wave dark matter masses. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:40:08 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Eberhardt",
"Andrew",
""
],
[
"Hui",
"Lam",
""
]
] |
2506.02401 | Chi Ding | Chi Ding, Junxiao Xue, Cong Wang, Hao Zhou | Trusted Fake Audio Detection Based on Dirichlet Distribution | null | null | null | null | cs.SD cs.MM eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | With the continuous development of deep learning-based speech conversion and speech synthesis technologies, the cybersecurity problem posed by fake audio has become increasingly serious. Previously proposed models for defending against fake audio have attained remarkable performance. However, they all fall short in modeling the trustworthiness of the decisions made by the models themselves. Based on this, we put forward a plausible fake audio detection approach based on the Dirichlet distribution with the aim of enhancing the reliability of fake audio detection. Specifically, we first generate evidence through a neural network. Uncertainty is then modeled using the Dirichlet distribution. By modeling the belief distribution with the parameters of the Dirichlet distribution, an estimate of uncertainty can be obtained for each decision. Finally, the predicted probabilities and corresponding uncertainty estimates are combined to form the final opinion. On the ASVspoof series dataset (i.e., ASVspoof 2019 LA, ASVspoof 2021 LA, and DF), we conduct a number of comparison experiments to verify the excellent performance of the proposed model in terms of accuracy, robustness, and trustworthiness. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:40:39 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Ding",
"Chi",
""
],
[
"Xue",
"Junxiao",
""
],
[
"Wang",
"Cong",
""
],
[
"Zhou",
"Hao",
""
]
] |
2506.02402 | Chao Zhao | Shujie Pan, Victoria Cao, Yiheng Feng, Dingyi Wu, Jie Yan, Junjie Yang, Chao Zhao, Xi Xiao, and Siming Chen | Highly reliable, ultra-wideband, isolator-free quantum-dot mode-locked frequency combs for optical interconnects beyond 3.2Tb/s | null | null | null | null | cond-mat.mes-hall physics.optics | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum dot mode-locked laser-based optical frequency combs are emerging as a critical solution for achieving low-cost, high-efficiency, and large-capacity optical interconnects. The practical implementation of wavelength division multiplexing interconnects necessitates a temperature-stable OFC source with a minimum 100 GHz channel spacing to enable high-bandwidth modulation while mitigating the complexity of optical filtering and detection. By leveraging the advanced co-doping technique and a colliding pulse mode-locking scheme, here, we report a compact, ultra-wideband, highly reliable, isolator-free 100 GHz-spacing InAs/GaAs QD OFC source operating up to a record temperature of 140 {\deg}C. The comb source delivers a record 3 dB optical bandwidth of 14.312 nm, containing flat-top comb lines, each supporting 128 Gb/s PAM-4 modulation, which results in a total throughput of 3.328 Tb/s with an extremely low power consumption of 0.394 pJ/bit at 25{\deg}C. Performance remains stable at 85 {\deg}C, with negligible degradation of device critical metrics. Remarkably, accelerated aging tests under harsh conditions (85 {\deg}C with 8x threshold current injection) revealed a mean time to failure of approximately 207 years. The QD OFC source demonstrated in this work, for the first time, establishes a concrete link between fundamental research on comb sources and their practical deployment in next-generation, high-density optical interconnect systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:43:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Pan",
"Shujie",
""
],
[
"Cao",
"Victoria",
""
],
[
"Feng",
"Yiheng",
""
],
[
"Wu",
"Dingyi",
""
],
[
"Yan",
"Jie",
""
],
[
"Yang",
"Junjie",
""
],
[
"Zhao",
"Chao",
""
],
[
"Xiao",
"Xi",
""
],
[
"Chen",
"Siming",
""
]
] |
2506.02403 | Kazuhiro Takemoto | Yasutoshi Nagahama, Kosuke Miyazato, Kazuhiro Takemoto | Adversarial control of synchronization in complex oscillator networks | 9 pages, 4 figures | null | null | null | nlin.AO physics.soc-ph | http://creativecommons.org/licenses/by/4.0/ | This study investigates adversarial attacks, a concept from deep learning, designed to control synchronization dynamics through strategically crafted minimal perturbations. We propose a gradient-based optimization method that identifies small phase perturbations to dramatically enhance or suppress collective synchronization in Kuramoto oscillator networks. Our approach formulates synchronization control as an adversarial optimization problem, computing gradients of the order parameter with respect to oscillator phases to determine optimal perturbation directions. Results demonstrate that extremely small phase perturbations applied to network oscillators can achieve significant synchronization control across diverse network architectures. Our analysis reveals that synchronization enhancement is achievable across various network sizes, while synchronization suppression becomes particularly effective in larger networks, with effectiveness scaling favorably with network size. The method is systematically validated on canonical model networks including scale-free and small-world topologies, and real-world networks representing power grids and brain connectivity patterns. This adversarial framework represents a novel paradigm for synchronization management by introducing deep learning concepts to networked dynamical systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:44:18 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Nagahama",
"Yasutoshi",
""
],
[
"Miyazato",
"Kosuke",
""
],
[
"Takemoto",
"Kazuhiro",
""
]
] |
2506.02404 | Yilin Xiao | Yilin Xiao, Junnan Dong, Chuang Zhou, Su Dong, Qianwen Zhang, Di Yin, Xing Sun, Xiao Huang | GraphRAG-Bench: Challenging Domain-Specific Reasoning for Evaluating Graph Retrieval-Augmented Generation | null | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Graph Retrieval Augmented Generation (GraphRAG) has garnered increasing recognition for its potential to enhance large language models (LLMs) by structurally organizing domain-specific corpora and facilitating complex reasoning. However, current evaluations of GraphRAG models predominantly rely on traditional question-answering datasets. Their limited scope in questions and evaluation metrics fails to comprehensively assess the reasoning capacity improvements enabled by GraphRAG models. To address this gap, we introduce GraphRAG-Bench, a large-scale, domain-specific benchmark designed to rigorously evaluate GraphRAG models. Our benchmark offers three key superiorities: \((i)\) Challenging question design. Featuring college-level, domain-specific questions that demand multi-hop reasoning, the benchmark ensures that simple content retrieval is insufficient for problem-solving. For example, some questions require mathematical reasoning or programming. \((ii)\) Diverse task coverage. The dataset includes a broad spectrum of reasoning tasks, multiple-choice, true/false, multi-select, open-ended, and fill-in-the-blank. It spans 16 disciplines in twenty core textbooks. \((iii)\) Holistic evaluation framework. GraphRAG-Bench provides comprehensive assessment across the entire GraphRAG pipeline, including graph construction, knowledge retrieval, and answer generation. Beyond final-answer correctness, it evaluates the logical coherence of the reasoning process. By applying nine contemporary GraphRAG methods to GraphRAG-Bench, we demonstrate its utility in quantifying how graph-based structuring improves model reasoning capabilities. Our analysis reveals critical insights about graph architectures, retrieval efficacy, and reasoning capabilities, offering actionable guidance for the research community. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:44:26 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Xiao",
"Yilin",
""
],
[
"Dong",
"Junnan",
""
],
[
"Zhou",
"Chuang",
""
],
[
"Dong",
"Su",
""
],
[
"Zhang",
"Qianwen",
""
],
[
"Yin",
"Di",
""
],
[
"Sun",
"Xing",
""
],
[
"Huang",
"Xiao",
""
]
] |
2506.02405 | Zhiya Tan | Zhiya Tan, Xin Zhang, Joey Tianyi Zhou | Modelship Attribution: Tracing Multi-Stage Manipulations Across Generative Models | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | As generative techniques become increasingly accessible, authentic visuals are frequently subjected to iterative alterations by various individuals employing a variety of tools. Currently, to avoid misinformation and ensure accountability, a lot of research on detection and attribution is emerging. Although these methods demonstrate promise in single-stage manipulation scenarios, they fall short when addressing complex real-world iterative manipulation. In this paper, we are the first, to the best of our knowledge, to systematically model this real-world challenge and introduce a novel method to solve it. We define a task called "Modelship Attribution", which aims to trace the evolution of manipulated images by identifying the generative models involved and reconstructing the sequence of edits they performed. To realistically simulate this scenario, we utilize three generative models, StyleMapGAN, DiffSwap, and FacePartsSwap, that sequentially modify distinct regions of the same image. This process leads to the creation of the first modelship dataset, comprising 83,700 images (16,740 images*5). Given that later edits often overwrite the fingerprints of earlier models, the focus shifts from extracting blended fingerprints to characterizing each model's distinctive editing patterns. To tackle this challenge, we introduce the modelship attribution transformer (MAT), a purpose-built framework designed to effectively recognize and attribute the contributions of various models within complex, multi-stage manipulation workflows. Through extensive experiments and comparative analysis with other related methods, our results, including comprehensive ablation studies, demonstrate that the proposed approach is a highly effective solution for modelship attribution. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:45:09 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tan",
"Zhiya",
""
],
[
"Zhang",
"Xin",
""
],
[
"Zhou",
"Joey Tianyi",
""
]
] |
2506.02406 | Renat Sergazinov | Renat Sergazinov and Jing Wu and Shao-An Yin | Random at First, Fast at Last: NTK-Guided Fourier Pre-Processing for Tabular DL | 16 pages, 3 figures, 1 table | null | null | null | cs.LG cs.AI stat.ML | http://creativecommons.org/licenses/by/4.0/ | While random Fourier features are a classic tool in kernel methods, their utility as a pre-processing step for deep learning on tabular data has been largely overlooked. Motivated by shortcomings in tabular deep learning pipelines - revealed through Neural Tangent Kernel (NTK) analysis - we revisit and repurpose random Fourier mappings as a parameter-free, architecture-agnostic transformation. By projecting each input into a fixed feature space via sine and cosine projections with frequencies drawn once at initialization, this approach circumvents the need for ad hoc normalization or additional learnable embeddings. We show within the NTK framework that this mapping (i) bounds and conditions the network's initial NTK spectrum, and (ii) introduces a bias that shortens the optimization trajectory, thereby accelerating gradient-based training. These effects pre-condition the network with a stable kernel from the outset. Empirically, we demonstrate that deep networks trained on Fourier-transformed inputs converge more rapidly and consistently achieve strong final performance, often with fewer epochs and less hyperparameter tuning. Our findings establish random Fourier pre-processing as a theoretically motivated, plug-and-play enhancement for tabular deep learning. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:45:13 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Sergazinov",
"Renat",
""
],
[
"Wu",
"Jing",
""
],
[
"Yin",
"Shao-An",
""
]
] |
2506.02407 | Qian Yin | Zhonghua Qiao, Zhenli Xu, Qian Yin, Shenggao Zhou | Intrinsic local Gauss's law preserving PIC method: A self-consistent field-particle update scheme for plasma simulations | null | null | null | null | physics.plasm-ph physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | In order to perform physically faithful particle-in-cell (PIC) simulations, the Gauss's law stands as a critical requirement, since its violation often leads to catastrophic errors in long-term plasma simulations. This work proposes a novel method that intrinsically enforces the Gauss's law for the Vlasov-Amp\`ere/Vlasov-Poisson system without requiring auxiliary field corrections or specialized current deposition techniques. The electric field is managed to get updated locally and consistently with the motion of particles via splitting the motion into sub-steps along each dimension of the computational mesh. To further obtain a curl-free electric field, a local update scheme is developed to relax the electric-field free energy subject to the Gauss's law. The proposed method avoids solving the Poisson's or Amp\`ere's equation, resulting in a local algorithm of linear complexity for each time step which can be flexibly combined with various temporal discretization for particle motion in PIC simulations. Theoretical analysis verifies that the proposed method indeed maintains the discrete Gauss's law exactly. Numerical tests on classical benchmarks, including the Landau damping, two-stream instability and Diocotron instability, demonstrate the key advantages of the proposed method. It is expected that the local nature of the proposed method makes it a promising tool in parallel simulations of large-scale plasmas. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:45:36 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Qiao",
"Zhonghua",
""
],
[
"Xu",
"Zhenli",
""
],
[
"Yin",
"Qian",
""
],
[
"Zhou",
"Shenggao",
""
]
] |
2506.02408 | Wenhao Tang | Wenhao Tang, Rong Qin, Heng Fang, Fengtao Zhou, Hao Chen, Xiang Li, Ming-Ming Cheng | Revisiting End-to-End Learning with Slide-level Supervision in Computational Pathology | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Pre-trained encoders for offline feature extraction followed by multiple instance learning (MIL) aggregators have become the dominant paradigm in computational pathology (CPath), benefiting cancer diagnosis and prognosis. However, performance limitations arise from the absence of encoder fine-tuning for downstream tasks and disjoint optimization with MIL. While slide-level supervised end-to-end (E2E) learning is an intuitive solution to this issue, it faces challenges such as high computational demands and suboptimal results. These limitations motivate us to revisit E2E learning. We argue that prior work neglects inherent E2E optimization challenges, leading to performance disparities compared to traditional two-stage methods. In this paper, we pioneer the elucidation of optimization challenge caused by sparse-attention MIL and propose a novel MIL called ABMILX. It mitigates this problem through global correlation-based attention refinement and multi-head mechanisms. With the efficient multi-scale random patch sampling strategy, an E2E trained ResNet with ABMILX surpasses SOTA foundation models under the two-stage paradigm across multiple challenging benchmarks, while remaining computationally efficient (<10 RTX3090 hours). We show the potential of E2E learning in CPath and calls for greater research focus in this area. The code is https://github.com/DearCaat/E2E-WSI-ABMILX. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:46:50 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tang",
"Wenhao",
""
],
[
"Qin",
"Rong",
""
],
[
"Fang",
"Heng",
""
],
[
"Zhou",
"Fengtao",
""
],
[
"Chen",
"Hao",
""
],
[
"Li",
"Xiang",
""
],
[
"Cheng",
"Ming-Ming",
""
]
] |
2506.02409 | Zeshan Haider | Zeshan Haider, Muhammad Altaf, Tahira Nasreen, Muhammad Imran, Rameez Ul Islam, Manzoor Ikram | The Observation of hyperradiance accompanied by enhanced entanglement in a hybrid optomechanical system | null | null | 10.1140/epjp/s13360-024-05073-7 | null | quant-ph | http://creativecommons.org/publicdomain/zero/1.0/ | We have theoretically investigated an optomechanical system and presented the scenario of significantly enhanced bipartite photon-phonon entanglement for two qubits coupled to the single mode of the cavity. And results are compared with the one qubit case for reference. The tripartite atoms-photon-phonon interaction is considered as only three-body resonant interaction while the two-body actions are ignored under some potential approximations. Furthermore, we have studied the phenomenon of hyperradiance in which the well-known Dicke superradiant ($N^2$ scaling law) can be surpassed due to the inter-atomic correlations. Jointly, a parameter regime is explored to observe the entanglement of photon-phonon pairs and their hyperradiance simultaneously. As it is important to show that the generation of photons and phonons are antibunched, the equal time second-order correlation function $g^{(2)}(0)$ is characterized as witness. This system can be realized in Circuit Cavity Quantum Electrodynamics (CCQED) in which the direct coupling of the atom and mechanical resonator is possible.We have theoretically investigated an optomechanical system and presented the scenario of significantly enhanced bipartite photon-phonon entanglement for two qubits coupled to the single mode of the cavity. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:50:25 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Haider",
"Zeshan",
""
],
[
"Altaf",
"Muhammad",
""
],
[
"Nasreen",
"Tahira",
""
],
[
"Imran",
"Muhammad",
""
],
[
"Islam",
"Rameez Ul",
""
],
[
"Ikram",
"Manzoor",
""
]
] |
2506.02410 | Shiwei Sang | Shiwei Sang, Yicheng Zeng, Xuehu Zhu, Shurong Zheng | Testing for large-dimensional covariance matrix under differential privacy | null | null | null | null | stat.ME | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The increasing prevalence of high-dimensional data across various applications has raised significant privacy concerns in statistical inference. In this paper, we propose a differentially private integrated statistic for testing large-dimensional covariance structures, enabling accurate statistical insights while safeguarding privacy. First, we analyze the global sensitivity of sample eigenvalues for sub-Gaussian populations, where our method bypasses the commonly assumed boundedness of data covariates. For sufficiently large sample size, the privatized statistic guarantees privacy with high probability. Furthermore, when the ratio of dimension to sample size, $d/n \to y \in (0, \infty)$, the privatized test is asymptotically distribution-free with well-known critical values, and detects the local alternative hypotheses distinct from the null at the fastest rate of $1/\sqrt{n}$. Extensive numerical studies on synthetic and real data showcase the validity and powerfulness of our proposed method. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:53:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Sang",
"Shiwei",
""
],
[
"Zeng",
"Yicheng",
""
],
[
"Zhu",
"Xuehu",
""
],
[
"Zheng",
"Shurong",
""
]
] |
2506.02411 | Xiaokun Teng | Xiaokun Teng, Wankai Tang, Xiao Li, Shi Jin | Baseband-Free End-to-End Communication System Based on Diffractive Deep Neural Network | submitted to IEEE journal for possible publication | null | null | null | eess.SP cs.IT math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffractive deep neural network (D2NN), also referred to as reconfigurable intelligent metasurface based deep neural networks (Rb-DNNs) or stacked intelligent metasurfaces (SIMs) in the field of wireless communications, has emerged as a promising signal processing paradigm that enables computing-by-propagation. However, existing architectures are limited to implementing specific functions such as precoding and combining, while still relying on digital baseband modules for other essential tasks like modulation and detection. In this work, we propose a baseband-free end-to-end (BBF-E2E) wireless communication system where modulation, beamforming, and detection are jointly realized through the propagation of electromagnetic (EM) waves. The BBF-E2E system employs D2NNs at both the transmitter and the receiver, forming an autoencoder architecture optimized as a complex-valued neural network. The transmission coefficients of each metasurface layer are trained using the mini-batch stochastic gradient descent method to minimize the cross-entropy loss. To reduce computational complexity during diffraction calculation, the angular spectrum method (ASM) is adopted in place of the Rayleigh-Sommerfeld formula. Extensive simulations demonstrate that BBF-E2E achieves robust symbol transmission under challenging channel conditions with significantly reduced hardware requirements. In particular, the proposed system matches the performance of a conventional multi-antenna system with 81 RF chains while requiring only a single RF chain and 1024 passive elements of metasurfaces. These results highlight the potential of wave-domain neural computing to replace digital baseband modules in future wireless transceivers. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:56:09 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Teng",
"Xiaokun",
""
],
[
"Tang",
"Wankai",
""
],
[
"Li",
"Xiao",
""
],
[
"Jin",
"Shi",
""
]
] |
2506.02412 | Zhengyuan Liu | Zhengyuan Liu, Geyu Lin, Hui Li Tan, Huayun Zhang, Yanfeng Lu, Xiaoxue Gao, Stella Xin Yin, He Sun, Hock Huan Goh, Lung Hsiang Wong, Nancy F. Chen | SingaKids: A Multilingual Multimodal Dialogic Tutor for Language Learning | ACL 2025 Industry Track | null | null | null | cs.CL cs.AI | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The integration of generative artificial intelligence into educational applications has enhanced personalized and interactive learning experiences, and it shows strong potential to promote young learners language acquisition. However, it is still challenging to ensure consistent and robust performance across different languages and cultural contexts, and kids-friendly design requires simplified instructions, engaging interactions, and age-appropriate scaffolding to maintain motivation and optimize learning outcomes. In this work, we introduce SingaKids, a dialogic tutor designed to facilitate language learning through picture description tasks. Our system integrates dense image captioning, multilingual dialogic interaction, speech understanding, and engaging speech generation to create an immersive learning environment in four languages: English, Mandarin, Malay, and Tamil. We further improve the system through multilingual pre-training, task-specific tuning, and scaffolding optimization. Empirical studies with elementary school students demonstrate that SingaKids provides effective dialogic teaching, benefiting learners at different performance levels. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:56:45 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Zhengyuan",
""
],
[
"Lin",
"Geyu",
""
],
[
"Tan",
"Hui Li",
""
],
[
"Zhang",
"Huayun",
""
],
[
"Lu",
"Yanfeng",
""
],
[
"Gao",
"Xiaoxue",
""
],
[
"Yin",
"Stella Xin",
""
],
[
"Sun",
"He",
""
],
[
"Goh",
"Hock Huan",
""
],
[
"Wong",
"Lung Hsiang",
""
],
[
"Chen",
"Nancy F.",
""
]
] |
2506.02413 | Tian Lan | Tian Lan, Jie Guo and Chen Zhang | Tensor State Space-based Dynamic Multilayer Network Modeling | null | null | null | null | stat.ML cs.LG | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding the complex interactions within dynamic multilayer networks is critical for advancements in various scientific domains. Existing models often fail to capture such networks' temporal and cross-layer dynamics. This paper introduces a novel Tensor State Space Model for Dynamic Multilayer Networks (TSSDMN), utilizing a latent space model framework. TSSDMN employs a symmetric Tucker decomposition to represent latent node features, their interaction patterns, and layer transitions. Then by fixing the latent features and allowing the interaction patterns to evolve over time, TSSDMN uniquely captures both the temporal dynamics within layers and across different layers. The model identifiability conditions are discussed. By treating latent features as variables whose posterior distributions are approximated using a mean-field variational inference approach, a variational Expectation Maximization algorithm is developed for efficient model inference. Numerical simulations and case studies demonstrate the efficacy of TSSDMN for understanding dynamic multilayer networks. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 03:58:32 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Lan",
"Tian",
""
],
[
"Guo",
"Jie",
""
],
[
"Zhang",
"Chen",
""
]
] |
2506.02414 | Fengjin Li | Fengjin Li, Jie Wang, Yadong Niu, Yongqing Wang, Meng Meng, Jian Luan, Zhiyong Wu | StarVC: A Unified Auto-Regressive Framework for Joint Text and Speech Generation in Voice Conversion | 5 pages, 2 figures, Accepted by Interspeech 2025, Demo: https://thuhcsi.github.io/StarVC/ | null | null | null | cs.MM cs.CL cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Voice Conversion (VC) modifies speech to match a target speaker while preserving linguistic content. Traditional methods usually extract speaker information directly from speech while neglecting the explicit utilization of linguistic content. Since VC fundamentally involves disentangling speaker identity from linguistic content, leveraging structured semantic features could enhance conversion performance. However, previous attempts to incorporate semantic features into VC have shown limited effectiveness, motivating the integration of explicit text modeling. We propose StarVC, a unified autoregressive VC framework that first predicts text tokens before synthesizing acoustic features. The experiments demonstrate that StarVC outperforms conventional VC methods in preserving both linguistic content (i.e., WER and CER) and speaker characteristics (i.e., SECS and MOS). Audio demo can be found at: https://thuhcsi.github.io/StarVC/. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:00:53 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Fengjin",
""
],
[
"Wang",
"Jie",
""
],
[
"Niu",
"Yadong",
""
],
[
"Wang",
"Yongqing",
""
],
[
"Meng",
"Meng",
""
],
[
"Luan",
"Jian",
""
],
[
"Wu",
"Zhiyong",
""
]
] |
2506.02415 | Karthikeyan Vaiapury | Karthikeyan Vaiapury | AERO: A Redirection-Based Optimization Framework Inspired by Judo for Robust Probabilistic Forecasting | 15 pages, 1 figure, submitted to NeurIPS 2025 (preprint version) | null | null | null | cs.LG cs.AI | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Optimization remains a fundamental pillar of machine learning, yet existing methods often struggle to maintain stability and adaptability in dynamic, non linear systems, especially under uncertainty. We introduce AERO (Adversarial Energy-based Redirection Optimization), a novel framework inspired by the redirection principle in Judo, where external disturbances are leveraged rather than resisted. AERO reimagines optimization as a redirection process guided by 15 interrelated axioms encompassing adversarial correction, energy conservation, and disturbance-aware learning. By projecting gradients, integrating uncertainty driven dynamics, and managing learning energy, AERO offers a principled approach to stable and robust model updates. Applied to probabilistic solar energy forecasting, AERO demonstrates substantial gains in predictive accuracy, reliability, and adaptability, especially in noisy and uncertain environments. Our findings highlight AERO as a compelling new direction in the theoretical and practical landscape of optimization. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:02:20 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Vaiapury",
"Karthikeyan",
""
]
] |
2506.02416 | Mikhail Kaygorodov | M. Y. Kaygorodov, Y. S. Kozhedub, A. V. Malyshev, A. O. Davydov, Y. Wu, S. B. Zhang | Study of atomic effects on electron spectrum in bound-muon decay process | 16 pages, 8 figures, 1 table | null | null | null | physics.atom-ph hep-ph | http://creativecommons.org/publicdomain/zero/1.0/ | For the bound-muon decay process, the study of atomic effects on the electron spectrum near its endpoint is performed within the framework of the Fermi effective theory. The analysis takes into account for corrections due to finite-nuclear-size, nuclear-deformation, electron-screening, and vacuum-polarization effects, all of which are incorporated self-consistently into the Dirac equation. Furthermore, the nuclear-recoil correction to the muon binding energy is included. Calculations are carried out for the isotopes of C, Al, and Si, which are of a particular importance for forthcoming experiments aimed at search for the charged-lepton flavor-violating process of muon-to-electron conversion in a nuclear field. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:04:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kaygorodov",
"M. Y.",
""
],
[
"Kozhedub",
"Y. S.",
""
],
[
"Malyshev",
"A. V.",
""
],
[
"Davydov",
"A. O.",
""
],
[
"Wu",
"Y.",
""
],
[
"Zhang",
"S. B.",
""
]
] |
2506.02417 | Fujin Wang | Fujin Wang, Syed Masood, L.G. Wang | Quantum light and radiation in Rindler spacetime: from uncertainty relations to the cosmological implications | 11 pages, 6 figures, | null | null | null | gr-qc physics.optics | http://creativecommons.org/licenses/by/4.0/ | Based on an analogy between diffraction integral formalism of classical field propagation and Feynman path integral approach to quantum field theory, we develop a quantum model for light and radiation in Rindler spacetime. The framework helps to reveal acceleration-induced contributions to the traditional Heisenberg position-momentum uncertainty relation. A modified Planck energy density distribution of radiation is established and reveals equivalence between temperature and Rindler acceleration as advocated by standard Unruh and anti-Unruh effects. Later, by defining an equivalent acceleration, we investigate some cosmological implications of the model with regards to redshift and expansion of the Universe. In this context, we contend that the accelerated expansion of the Universe, in addition to possessing some well-defined limits corresponding to early and local Universe epochs, may also hint towards dynamical nature of dark energy. The findings provide glimpse into future table-top experiments aimed at emulating gravitational and other cosmological phenomena in terrestrial lab setups. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:05:21 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wang",
"Fujin",
""
],
[
"Masood",
"Syed",
""
],
[
"Wang",
"L. G.",
""
]
] |
2506.02418 | Wenxuan Pan | Wenxuan Pan, Yang Yang, Dong Wei, Meng Zhang, Zhiyu Zhu | Passive Multi-Target Visible Light Positioning Based on Multi-Camera Joint Optimization | null | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Camera-based visible light positioning (VLP) has emerged as a promising indoor positioning technique. However, the need for dedicated LED infrastructure and on-target cameras in existing algorithms limits their scalability and increases deployment costs. To address these limitations, this letter proposes a passive VLP algorithm based on Multi-Camera Joint Optimization (MCJO). In the considered system, multiple pre-calibrated cameras mounted on the ceiling continuously capture images of positioning targets equipped with point light sources, and can simultaneously localize these targets at the server. In particular, the proposed MCJO comprises two stages: It first estimates target positions via linear least squares from multi-view projection rays; then refines these positions through nonlinear joint optimization to minimize the reprojection error. Simulation results show that MCJO can achieve millimeter-level accuracy, with an improvement of 19% over state-of-the-art algorithms. Experimental results further show that MCJO can achieve an average position error as low as 5.63 mm. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:05:43 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Pan",
"Wenxuan",
""
],
[
"Yang",
"Yang",
""
],
[
"Wei",
"Dong",
""
],
[
"Zhang",
"Meng",
""
],
[
"Zhu",
"Zhiyu",
""
]
] |
2506.02419 | Nurislam Tursynbek | Nurislam Tursynbek, Hastings Greer, Basar Demir, Marc Niethammer | Guiding Registration with Emergent Similarity from Pre-Trained Diffusion Models | MICCAI 2025 | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Diffusion models, while trained for image generation, have emerged as powerful foundational feature extractors for downstream tasks. We find that off-the-shelf diffusion models, trained exclusively to generate natural RGB images, can identify semantically meaningful correspondences in medical images. Building on this observation, we propose to leverage diffusion model features as a similarity measure to guide deformable image registration networks. We show that common intensity-based similarity losses often fail in challenging scenarios, such as when certain anatomies are visible in one image but absent in another, leading to anatomically inaccurate alignments. In contrast, our method identifies true semantic correspondences, aligning meaningful structures while disregarding those not present across images. We demonstrate superior performance of our approach on two tasks: multimodal 2D registration (DXA to X-Ray) and monomodal 3D registration (brain-extracted to non-brain-extracted MRI). Code: https://github.com/uncbiag/dgir | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:07:04 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tursynbek",
"Nurislam",
""
],
[
"Greer",
"Hastings",
""
],
[
"Demir",
"Basar",
""
],
[
"Niethammer",
"Marc",
""
]
] |
2506.02420 | Yanyu Cheng | Yanyu Cheng, Chongjun Ouyang, Yuanwei Liu, George K. Karagiannidis | On the Performance of Pinching-Antenna Systems (PASS) with Orthogonal and Non-Orthogonal Multiple Access | null | null | null | null | eess.SP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper conducts a comprehensive performance analysis for pinching-antenna systems (PASS) under both orthogonal multiple access (OMA) and non-orthogonal multiple access (NOMA) transmission. Given the cost of waveguides, we consider a scenario where the waveguide is not deployed in all rooms, i.e., some users are beyond the line-of-sight (LoS) link service area of the PASS. Specifically, we consider a PASS where a pinching antenna in one room serves two users located in separate rooms. The wireless transmissions between the pinching antenna and the users are performed via LoS and non-line-of-sight (NLoS) links, respectively. Closed-form expressions for the outage probabilities (OPs) of the two users are derived for the considered system. Furthermore, asymptotic analyses in the high signal-to-noise ratio (SNR) regime are performed to reveal the achievable diversity orders. Numerical simulations validate the accuracy of the theoretical analysis and show that: 1) compared with conventional antenna systems (CASS), the OP of the LoS user in PASS is significantly reduced for both OMA and NOMA schemes in the middle SNR regime and approaches zero as the SNR increases; 2) since the diversity orders of the NLoS user in CASS and PASS are the same, the movement of the pinching antenna has no significant effect on the OP of the NLoS user for the OMA and NOMA scenarios. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:08:42 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cheng",
"Yanyu",
""
],
[
"Ouyang",
"Chongjun",
""
],
[
"Liu",
"Yuanwei",
""
],
[
"Karagiannidis",
"George K.",
""
]
] |
2506.02421 | Ivan Ovtin | KEDR Collaboration, V.V.Anashin, O.V. Anchugov, A.V. Andrianov, K.V. Astrelina, V.M. Aulchenko, E.M. Baldin, G.N. Baranov, A.K. Barladyan, A.Yu. Barnyakov, M.Yu. Barnyakov, S.E. Baru, I.Yu. Basok, A.M. Batrakov, I.V. Bedny, E.A. Bekhtenev, O.V. Belikov, D.E. Berkaev, A.E. Blinov, V.E. Blinov, M.F. Blinov, A.V. Bobrov, V.S. Bobrovnikov, A.V. Bogomyagkov, D. Yu. Bolkhovityanov, S.I. Eidelman, A.E. Bondar, D.V. Bondarev, A.R. Buzykaev, P.B. Cheblakov, V.L. Dorohov, F.A. Emanov, V.V. Gambaryan, Yu.M. Glukhovchenko, D.N. Grigoriev, V.V. Gulevich, D.V. Gusev, V.V. Kaminskiy, S.E. Karnaev, S.V. Karpov, G.V. Karpov, K.Yu. Karukina, D.P. Kashtankin, P.V. Kasyanenko, A.A. Katcin, T.A. Kharlamova, V.A. Kiselev, S.A. Kononov, K.Yu. Kotov, A.A. Krasnov, E.A. Kravchenko, V.N. Kudryavtsev, V.F. Kulikov, G.Ya. Kurkina, E.A. Kuper, I.A. Kuyanov, E.B. Levichev, P.V. Logachev, D.A. Maksimov, T.V. Maltsev, Yu.I. Maltseva, V.M. Malyshev, A.L. Maslennikov, A.S. Medvedko, O.I. Meshkov, S.I. Mishnev, I.A. Morozov, I.I. Morozov, N.Yu. Muchnoi, V.V. Neufeld, D.A. Nikiforov, S.A. Nikitin, I.B. Nikolaev, I.N. Okunev, A.P. Onuchin, S.B. Oreshkin, I.O. Orlov, A.A. Osipov, I.V. Ovtin, A.V. Pavlenko, S.V. Peleganchuk, K.G. Petrukhin, P. A. Piminov, S.G. Pivovarov, A.O. Poluektov, I.N. Popkov, V.G. Prisekin, O.L. Rezanova, A.A. Ruban, V.K. Sandyrev, G.A. Savinov, A.G. Shamov, E.A. Simonov, D.N. Shatilov, L.I. Shekhtman, D.A. Shvedov, B.A. Shwartz, Yu.I. Skovpen, A.N. Skrinsky, V.V. Smaluk, A.V. Sokolov, S.V. Sinyatkin, D.P. Sukhanov, A.M. Sukharev, A.A. Talyshev, E.V. Starostina, V.I. Telnov, Yu.A. Tikhonov, K.Yu. Todyshev, A.G. Tribendis, V.A. Tayursky, G.M. Tumaikin, Yu.V. Usov, A.I. Vorobiov, A.N. Yushkov, V.N. Zhilich, A.A. Zhukov, V.V. Zhulanova, A.N. Zhuravlev, D.A. Zubkov | New measurement of D0 and D+ meson masses with the KEDR detector | arXiv admin note: substantial text overlap with arXiv:0909.5545 | null | null | null | hep-ex | http://creativecommons.org/licenses/by/4.0/ | Using the 4.9 pb$^{-1}$ statistics collected at the peak of the $\psi(3770)$ resonance with the KEDR detector at the VEPP-4M electron-positron collider, we measured the masses of the neutral and charged D mesons: $M_{D^0} = 1865.100 \pm 0.210 (stat) \pm 0.046 (syst) MeV,$ and $M_{D^+} = 1869.560 \pm 0.288 (stat) \pm 0.109 (syst) MeV$. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:10:47 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"KEDR Collaboration",
"",
""
],
[
"Anashin",
"V. V.",
""
],
[
"Anchugov",
"O. V.",
""
],
[
"Andrianov",
"A. V.",
""
],
[
"Astrelina",
"K. V.",
""
],
[
"Aulchenko",
"V. M.",
""
],
[
"Baldin",
"E. M.",
""
],
[
"Baranov",
"G. N.",
""
],
[
"Barladyan",
"A. K.",
""
],
[
"Barnyakov",
"A. Yu.",
""
],
[
"Barnyakov",
"M. Yu.",
""
],
[
"Baru",
"S. E.",
""
],
[
"Basok",
"I. Yu.",
""
],
[
"Batrakov",
"A. M.",
""
],
[
"Bedny",
"I. V.",
""
],
[
"Bekhtenev",
"E. A.",
""
],
[
"Belikov",
"O. V.",
""
],
[
"Berkaev",
"D. E.",
""
],
[
"Blinov",
"A. E.",
""
],
[
"Blinov",
"V. E.",
""
],
[
"Blinov",
"M. F.",
""
],
[
"Bobrov",
"A. V.",
""
],
[
"Bobrovnikov",
"V. S.",
""
],
[
"Bogomyagkov",
"A. V.",
""
],
[
"Bolkhovityanov",
"D. Yu.",
""
],
[
"Eidelman",
"S. I.",
""
],
[
"Bondar",
"A. E.",
""
],
[
"Bondarev",
"D. V.",
""
],
[
"Buzykaev",
"A. R.",
""
],
[
"Cheblakov",
"P. B.",
""
],
[
"Dorohov",
"V. L.",
""
],
[
"Emanov",
"F. A.",
""
],
[
"Gambaryan",
"V. V.",
""
],
[
"Glukhovchenko",
"Yu. M.",
""
],
[
"Grigoriev",
"D. N.",
""
],
[
"Gulevich",
"V. V.",
""
],
[
"Gusev",
"D. V.",
""
],
[
"Kaminskiy",
"V. V.",
""
],
[
"Karnaev",
"S. E.",
""
],
[
"Karpov",
"S. V.",
""
],
[
"Karpov",
"G. V.",
""
],
[
"Karukina",
"K. Yu.",
""
],
[
"Kashtankin",
"D. P.",
""
],
[
"Kasyanenko",
"P. V.",
""
],
[
"Katcin",
"A. A.",
""
],
[
"Kharlamova",
"T. A.",
""
],
[
"Kiselev",
"V. A.",
""
],
[
"Kononov",
"S. A.",
""
],
[
"Kotov",
"K. Yu.",
""
],
[
"Krasnov",
"A. A.",
""
],
[
"Kravchenko",
"E. A.",
""
],
[
"Kudryavtsev",
"V. N.",
""
],
[
"Kulikov",
"V. F.",
""
],
[
"Kurkina",
"G. Ya.",
""
],
[
"Kuper",
"E. A.",
""
],
[
"Kuyanov",
"I. A.",
""
],
[
"Levichev",
"E. B.",
""
],
[
"Logachev",
"P. V.",
""
],
[
"Maksimov",
"D. A.",
""
],
[
"Maltsev",
"T. V.",
""
],
[
"Maltseva",
"Yu. I.",
""
],
[
"Malyshev",
"V. M.",
""
],
[
"Maslennikov",
"A. L.",
""
],
[
"Medvedko",
"A. S.",
""
],
[
"Meshkov",
"O. I.",
""
],
[
"Mishnev",
"S. I.",
""
],
[
"Morozov",
"I. A.",
""
],
[
"Morozov",
"I. I.",
""
],
[
"Muchnoi",
"N. Yu.",
""
],
[
"Neufeld",
"V. V.",
""
],
[
"Nikiforov",
"D. A.",
""
],
[
"Nikitin",
"S. A.",
""
],
[
"Nikolaev",
"I. B.",
""
],
[
"Okunev",
"I. N.",
""
],
[
"Onuchin",
"A. P.",
""
],
[
"Oreshkin",
"S. B.",
""
],
[
"Orlov",
"I. O.",
""
],
[
"Osipov",
"A. A.",
""
],
[
"Ovtin",
"I. V.",
""
],
[
"Pavlenko",
"A. V.",
""
],
[
"Peleganchuk",
"S. V.",
""
],
[
"Petrukhin",
"K. G.",
""
],
[
"Piminov",
"P. A.",
""
],
[
"Pivovarov",
"S. G.",
""
],
[
"Poluektov",
"A. O.",
""
],
[
"Popkov",
"I. N.",
""
],
[
"Prisekin",
"V. G.",
""
],
[
"Rezanova",
"O. L.",
""
],
[
"Ruban",
"A. A.",
""
],
[
"Sandyrev",
"V. K.",
""
],
[
"Savinov",
"G. A.",
""
],
[
"Shamov",
"A. G.",
""
],
[
"Simonov",
"E. A.",
""
],
[
"Shatilov",
"D. N.",
""
],
[
"Shekhtman",
"L. I.",
""
],
[
"Shvedov",
"D. A.",
""
],
[
"Shwartz",
"B. A.",
""
],
[
"Skovpen",
"Yu. I.",
""
],
[
"Skrinsky",
"A. N.",
""
],
[
"Smaluk",
"V. V.",
""
],
[
"Sokolov",
"A. V.",
""
],
[
"Sinyatkin",
"S. V.",
""
],
[
"Sukhanov",
"D. P.",
""
],
[
"Sukharev",
"A. M.",
""
],
[
"Talyshev",
"A. A.",
""
],
[
"Starostina",
"E. V.",
""
],
[
"Telnov",
"V. I.",
""
],
[
"Tikhonov",
"Yu. A.",
""
],
[
"Todyshev",
"K. Yu.",
""
],
[
"Tribendis",
"A. G.",
""
],
[
"Tayursky",
"V. A.",
""
],
[
"Tumaikin",
"G. M.",
""
],
[
"Usov",
"Yu. V.",
""
],
[
"Vorobiov",
"A. I.",
""
],
[
"Yushkov",
"A. N.",
""
],
[
"Zhilich",
"V. N.",
""
],
[
"Zhukov",
"A. A.",
""
],
[
"Zhulanova",
"V. V.",
""
],
[
"Zhuravlev",
"A. N.",
""
],
[
"Zubkov",
"D. A.",
""
]
] |
2506.02423 | Weicheng Zhan | Daomin Cao, Juncheng Wei, Weicheng Zhan | Symmetry in Serrin-type overdetermined problems | null | null | null | null | math.AP | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | This paper investigates the geometric constraints imposed on a domain by overdetermined problems for partial differential equations. Serrin's symmetry results are extended to overdetermined problems with potentially degenerate ellipticity in nonsmooth bounded domains. Furthermore, analogous symmetry results are established for ring-shaped domains. The proof relies on continuous Steiner symmetrization, along with a carefully constructed approximation argument. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:13:56 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cao",
"Daomin",
""
],
[
"Wei",
"Juncheng",
""
],
[
"Zhan",
"Weicheng",
""
]
] |
2506.02424 | Shukui Chen | Shukui Chen, Kirill Serkh, James Bremer, Murdock Aubry | An adaptive delaminating Levin method in two dimensions | null | null | null | null | math.NA cs.NA | http://creativecommons.org/licenses/by-nc-nd/4.0/ | We present an adaptive delaminating Levin method for evaluating bivariate oscillatory integrals over rectangular domains. Whereas previous analyses of Levin methods impose non-resonance conditions that exclude stationary and resonance points, we rigorously establish the existence of a slowly-varying, approximate solution to the Levin PDE across all frequency regimes, even when the non-resonance condition is violated. This allows us to derive error estimates for the numerical solution of the Levin PDE via the Chebyshev spectral collocation method, and for the evaluation of the corresponding oscillatory integrals, showing that high accuracy can be achieved regardless of whether or not stationary and resonance points are present. We then present a Levin method incorporating adaptive subdivision in both two and one dimensions, as well as delaminating Chebyshev spectral collocation, which is effective in both the presence and absence of stationary and resonance points. We demonstrate the effectiveness of our algorithm with a number of numerical experiments. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:14:52 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Shukui",
""
],
[
"Serkh",
"Kirill",
""
],
[
"Bremer",
"James",
""
],
[
"Aubry",
"Murdock",
""
]
] |
2506.02425 | Tairan Liu | Tairan Liu | Gender Inequality in English Textbooks Around the World: an NLP Approach | null | null | null | null | cs.CL stat.AP | http://creativecommons.org/licenses/by/4.0/ | Textbooks play a critical role in shaping children's understanding of the world. While previous studies have identified gender inequality in individual countries' textbooks, few have examined the issue cross-culturally. This study applies natural language processing methods to quantify gender inequality in English textbooks from 22 countries across 7 cultural spheres. Metrics include character count, firstness (which gender is mentioned first), and TF-IDF word associations by gender. The analysis also identifies gender patterns in proper names appearing in TF-IDF word lists, tests whether large language models can distinguish between gendered word lists, and uses GloVe embeddings to examine how closely keywords associate with each gender. Results show consistent overrepresentation of male characters in terms of count, firstness, and named entities. All regions exhibit gender inequality, with the Latin cultural sphere showing the least disparity. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:16:09 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liu",
"Tairan",
""
]
] |
2506.02428 | Victor Hugo Patty Yujra | Efrain Cruz-Mullisaca and Victor H. Patty-Yujra | Lie algebra rank condition for bilinear control systems on $\mathbb{R}^2$ | 13 pages, 5 figures | null | null | null | math.OC | http://creativecommons.org/publicdomain/zero/1.0/ | We will study the controllability problem of a bilinear control system on $\mathbb{R}^2:$ the main result is the characterization of the Lie algebra rank condition for the system. On the other hand, using elementary techniques, we recover conditions for the controllability of the induced angular system on the projective space. Finally, we will give controllability criteria for the system. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:26:08 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cruz-Mullisaca",
"Efrain",
""
],
[
"Patty-Yujra",
"Victor H.",
""
]
] |
2506.02429 | Luis Wirth | Luis Wirth | Rust Implementation of Finite Element Exterior Calculus on Coordinate-Free Simplicial Complexes | Bachelor's Thesis of Luis Wirth under the supervision of Prof. Dr. Ralf Hiptmair, for Computational Science and Engineering at ETH Z\"urich | null | null | null | math.NA cs.NA math.AP math.AT math.DG math.FA | http://creativecommons.org/licenses/by/4.0/ | This thesis presents the development of a novel finite element library in Rust based on the principles of Finite Element Exterior Calculus (FEEC). The library solves partial differential equations formulated using differential forms on abstract, coordinate-free simplicial complexes in arbitrary dimensions, employing an intrinsic Riemannian metric derived from edge lengths via Regge Calculus. We focus on solving elliptic Hodge-Laplace eigenvalue and source problems on the nD de Rham complex. We restrict ourselves to first-order Whitney basis functions. The implementation is partially verified through convergence studies. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:28:52 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wirth",
"Luis",
""
]
] |
2506.02430 | Teruaki Suyama | Teruaki Suyama and Shasvath J. Kapadia | On the Phase-Magnitude Relation in Gravitational Lensing: Reformulation and Applications of the Kramers-Kronig relation | 17 pages, 8 figures | null | null | null | gr-qc astro-ph.CO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | It is known that the amplification factor, defined as the ratio of the lensed to the unlensed waveform in the frequency domain, satisfies the Kramers-Kronig (KK) relation, which connects the real and imaginary parts of the amplification factor for any lensing signal. In this work, we reformulate the KK relation in terms of the magnitude and phase of the amplification factor. Unlike the original formulation, the phase cannot be uniquely determined from the magnitude alone due to the possible presence of a Blaschke product. While this ambiguity does not arise in the case of a point-mass lens, it can appear in more complex lens models, such as those with an NFW lens profile. As an application of our formulation, we demonstrate that the leading-order behavior of the phase in the low-frequency regime is completely determined by the leading-order behavior of the magnitude in the same regime. This reproduces known results from the literature, derived via low-frequency expansions for specific lens models. Importantly, our result does not rely on any particular lens model, highlighting a universal feature that the low-frequency behavior of the amplification factor is tightly constrained by the KK relation. As a further application, we present two examples in which the phase is constructed from a given analytic form of the magnitude using the newly derived KK relation. In particular, the second example allows for an analytic evaluation of the KK integral, yielding an explicit expression for the phase. This study offers a potentially powerful method for applying the KK relation in model-agnostic searches for lensing signals. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:33:32 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Suyama",
"Teruaki",
""
],
[
"Kapadia",
"Shasvath J.",
""
]
] |
2506.02431 | Mahammed Kamruzzaman | Mahammed Kamruzzaman, Abdullah Al Monsur, Gene Louis Kim, Anshuman Chhabra | From Anger to Joy: How Nationality Personas Shape Emotion Attribution in Large Language Models | null | null | null | null | cs.CL | http://creativecommons.org/licenses/by/4.0/ | Emotions are a fundamental facet of human experience, varying across individuals, cultural contexts, and nationalities. Given the recent success of Large Language Models (LLMs) as role-playing agents, we examine whether LLMs exhibit emotional stereotypes when assigned nationality-specific personas. Specifically, we investigate how different countries are represented in pre-trained LLMs through emotion attributions and whether these attributions align with cultural norms. Our analysis reveals significant nationality-based differences, with emotions such as shame, fear, and joy being disproportionately assigned across regions. Furthermore, we observe notable misalignment between LLM-generated and human emotional responses, particularly for negative emotions, highlighting the presence of reductive and potentially biased stereotypes in LLM outputs. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:35:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Kamruzzaman",
"Mahammed",
""
],
[
"Monsur",
"Abdullah Al",
""
],
[
"Kim",
"Gene Louis",
""
],
[
"Chhabra",
"Anshuman",
""
]
] |
2506.02432 | Sourabhashis Das | Sourabhashis Das, Wentang Kuo, and Yu-Ru Liu | Generalizations of Erd\H{o}s-Kac theorem with applications | Submitted for publication, 38 pages. arXiv admin note: text overlap with arXiv:2506.01030 | null | null | null | math.NT math.PR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Let $\omega(n)$ denote the number of distinct prime factors of a natural number $n$. In 1940, Erd\H{o}s and Kac established that $\omega(n)$ obeys the Gaussian distribution over natural numbers, and in 2004, the third author generalized their theorem to all abelian monoids. In this paper, we extend her theorem to any subsets of an abelian monoid satisfying some additional conditions, and apply this result to the subsets of $h$-free and $h$-full elements. We study generalizations of several arithmetic functions, such as the prime counting omega functions and the divisor function in a unified framework. Finally, we apply our results to number fields, global function fields, and geometrically irreducible projective varieties, demonstrating the broad relevance of our approach. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:36:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Das",
"Sourabhashis",
""
],
[
"Kuo",
"Wentang",
""
],
[
"Liu",
"Yu-Ru",
""
]
] |
2506.02433 | Shuqiang Wang | Weiheng Yao, Xuhang Chen, Shuqiang Wang | Empowering Functional Neuroimaging: A Pre-trained Generative Framework for Unified Representation of Neural Signals | null | null | null | null | cs.CV | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Multimodal functional neuroimaging enables systematic analysis of brain mechanisms and provides discriminative representations for brain-computer interface (BCI) decoding. However, its acquisition is constrained by high costs and feasibility limitations. Moreover, underrepresentation of specific groups undermines fairness of BCI decoding model. To address these challenges, we propose a unified representation framework for multimodal functional neuroimaging via generative artificial intelligence (AI). By mapping multimodal functional neuroimaging into a unified representation space, the proposed framework is capable of generating data for acquisition-constrained modalities and underrepresented groups. Experiments show that the framework can generate data consistent with real brain activity patterns, provide insights into brain mechanisms, and improve performance on downstream tasks. More importantly, it can enhance model fairness by augmenting data for underrepresented groups. Overall, the framework offers a new paradigm for decreasing the cost of acquiring multimodal functional neuroimages and enhancing the fairness of BCI decoding models. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:38:36 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yao",
"Weiheng",
""
],
[
"Chen",
"Xuhang",
""
],
[
"Wang",
"Shuqiang",
""
]
] |
2506.02435 | Zhen Zhang | Zhen Zhang, Luowen Liu, Wanzhi Zhang, Zitian Guo, Kun Huang, Qi Qi, Qiang Liu, Xingxing Wang | A Transformer-Based Neural Network for Optimal Deterministic-Allocation and Anonymous Joint Auction Design | null | null | null | null | cs.GT | http://creativecommons.org/licenses/by/4.0/ | With the advancement of machine learning, an increasing number of studies are employing automated mechanism design (AMD) methods for optimal auction design. However, all previous AMD architectures designed to generate optimal mechanisms that satisfy near dominant strategy incentive compatibility (DSIC) fail to achieve deterministic allocation, and some also lack anonymity, thereby impacting the efficiency and fairness of advertising allocation. This has resulted in a notable discrepancy between the previous AMD architectures for generating near-DSIC optimal mechanisms and the demands of real-world advertising scenarios. In this paper, we prove that in all online advertising scenarios, previous non-deterministic allocation methods lead to the non-existence of feasible solutions, resulting in a gap between the rounded solution and the optimal solution. Furthermore, we propose JTransNet, a transformer-based neural network architecture, designed for optimal deterministic-allocation and anonymous joint auction design. Although the deterministic allocation module in JTransNet is designed for the latest joint auction scenarios, it can be applied to other non-deterministic AMD architectures with minor modifications. Additionally, our offline and online data experiments demonstrate that, in joint auction scenarios, JTransNet significantly outperforms baseline methods in terms of platform revenue, resulting in a substantial increase in platform earnings. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:41:13 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhang",
"Zhen",
""
],
[
"Liu",
"Luowen",
""
],
[
"Zhang",
"Wanzhi",
""
],
[
"Guo",
"Zitian",
""
],
[
"Huang",
"Kun",
""
],
[
"Qi",
"Qi",
""
],
[
"Liu",
"Qiang",
""
],
[
"Wang",
"Xingxing",
""
]
] |
2506.02436 | Xueyan Zhu | Xueyan Zhu, Jin Yang, J. Shiomi, Cheng Shao | Revisiting the Mechanisms of Thermal Transport in Vacancy-Defective Silicon | null | null | null | null | cond-mat.mtrl-sci | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Understanding heat conduction in defective silicon is crucial for electronics and thermoelectrics. Conventional understanding relies on phonon gas picture, treating defects as scattering centers that reduce phonon lifetimes without altering frequencies and group velocities. We go beyond phonon gas picture by employing Wigner transport equation to investigate heat conduction in vacancy-defected silicon. Our findings reveal that while thermal conduction in pristine silicon stems mainly from particle-like propagation of vibrational modes, wave-like tunnelling becomes increasingly significant in the presence of vacancies. Contrary to the conventional belief that defects only perturb mode lifetimes, we demonstrate that vacancies also diminish velocity operators, a dominant factor in thermal conductivity reduction, surpassing even the effect of lifetime shortening. Furthermore, incorporating anharmonic frequencies and interatomic force constants shows that while anharmonicity suppresses thermal conductivity in pristine silicon, this effect weakens with vacancy concentration and reverses to enhance conductivity. These findings challenge conventional knowledge and provide new insights into thermal conduction in defective materials. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:42:41 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Zhu",
"Xueyan",
""
],
[
"Yang",
"Jin",
""
],
[
"Shiomi",
"J.",
""
],
[
"Shao",
"Cheng",
""
]
] |
2506.02437 | Peder Thompson | Petter Andreas Bergh, David A. Jorgensen, Peder Thompson | Multiplicity in triangulated categories | 24 pages; comments welcome! | null | null | null | math.KT math.AC math.CT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | We lay out the theory of a multiplicity in the setting of a triangulated category having a central ring action from a graded-commutative ring $R$, in other words, an $R$-linear triangulated category. The invariant we consider is modelled on those for graded modules over a commutative graded ring. We show that this invariant is determined by the leading coefficients of the Hilbert polynomials expressing the lengths of certain Hom sets. Our theory is a natural analogue of Hochster's theta invariant for homology and Buchweitz's Herbrand difference for cohomology. Moreover, we give applications to vanishing of cohomology and modules over local complete intersection rings, group algebras of a finite group, and certain finite dimensional algebras. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:45:10 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bergh",
"Petter Andreas",
""
],
[
"Jorgensen",
"David A.",
""
],
[
"Thompson",
"Peder",
""
]
] |
2506.02438 | Sudhanshu Sekhar Tripathy | Sudhanshu Sekhar Tripathy, Bichitrananda Behera | A Review of Various Datasets for Machine Learning Algorithm-Based Intrusion Detection System: Advances and Challenges | null | null | null | null | cs.CR cs.AI cs.LG | http://creativecommons.org/licenses/by/4.0/ | IDS aims to protect computer networks from security threats by detecting, notifying, and taking appropriate action to prevent illegal access and protect confidential information. As the globe becomes increasingly dependent on technology and automated processes, ensuring secured systems, applications, and networks has become one of the most significant problems of this era. The global web and digital technology have significantly accelerated the evolution of the modern world, necessitating the use of telecommunications and data transfer platforms. Researchers are enhancing the effectiveness of IDS by incorporating popular datasets into machine learning algorithms. IDS, equipped with machine learning classifiers, enhances security attack detection accuracy by identifying normal or abnormal network traffic. This paper explores the methods of capturing and reviewing intrusion detection systems (IDS) and evaluates the challenges existing datasets face. A deluge of research on machine learning (ML) and deep learning (DL) architecture-based intrusion detection techniques has been conducted in the past ten years on various cybersecurity datasets, including KDDCUP'99, NSL-KDD, UNSW-NB15, CICIDS-2017, and CSE-CIC-IDS2018. We conducted a literature review and presented an in-depth analysis of various intrusion detection methods that use SVM, KNN, DT, LR, NB, RF, XGBOOST, Adaboost, and ANN. We provide an overview of each technique, explaining the role of the classifiers and algorithms used. A detailed tabular analysis highlights the datasets used, classifiers employed, attacks detected, evaluation metrics, and conclusions drawn. This article offers a thorough review for future IDS research. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:47:21 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tripathy",
"Sudhanshu Sekhar",
""
],
[
"Behera",
"Bichitrananda",
""
]
] |
2506.02439 | Shuang Li | Shuang Li, Jiaxu Leng, Changjiang Kuang, Mingpi Tan, Xinbo Gao | Video-Level Language-Driven Video-Based Visible-Infrared Person Re-Identification | Accepted by IEEE TIFS | null | null | null | cs.CV | http://creativecommons.org/licenses/by/4.0/ | Video-based Visible-Infrared Person Re-Identification (VVI-ReID) aims to match pedestrian sequences across modalities by extracting modality-invariant sequence-level features. As a high-level semantic representation, language provides a consistent description of pedestrian characteristics in both infrared and visible modalities. Leveraging the Contrastive Language-Image Pre-training (CLIP) model to generate video-level language prompts and guide the learning of modality-invariant sequence-level features is theoretically feasible. However, the challenge of generating and utilizing modality-shared video-level language prompts to address modality gaps remains a critical problem. To address this problem, we propose a simple yet powerful framework, video-level language-driven VVI-ReID (VLD), which consists of two core modules: invariant-modality language prompting (IMLP) and spatial-temporal prompting (STP). IMLP employs a joint fine-tuning strategy for the visual encoder and the prompt learner to effectively generate modality-shared text prompts and align them with visual features from different modalities in CLIP's multimodal space, thereby mitigating modality differences. Additionally, STP models spatiotemporal information through two submodules, the spatial-temporal hub (STH) and spatial-temporal aggregation (STA), which further enhance IMLP by incorporating spatiotemporal information into text prompts. The STH aggregates and diffuses spatiotemporal information into the [CLS] token of each frame across the vision transformer (ViT) layers, whereas STA introduces dedicated identity-level loss and specialized multihead attention to ensure that the STH focuses on identity-relevant spatiotemporal feature aggregation. The VLD framework achieves state-of-the-art results on two VVI-ReID benchmarks. The code will be released at https://github.com/Visuang/VLD. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:49:08 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Li",
"Shuang",
""
],
[
"Leng",
"Jiaxu",
""
],
[
"Kuang",
"Changjiang",
""
],
[
"Tan",
"Mingpi",
""
],
[
"Gao",
"Xinbo",
""
]
] |
2506.02440 | Jamie Taylor | Jamie Taylor, Matija \v{C}ufar, David Mitrouskas, Robert Seiringer, Elke Pahl, Joachim Brand | Bound excited states of Fr\"ohlich polarons in one dimension | 11 pages, 9 figures | null | null | null | math-ph cond-mat.other math.MP physics.comp-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | The one-dimensional Fr\"ohlich model describing the motion of a single electron interacting with optical phonons is a paradigmatic model of quantum many-body physics. We predict the existence of an arbitrarily large number of bound excited states in the strong coupling limit and calculate their excitation energies. Numerical simulations of a discretized model demonstrate the complete amelioration of the projector Monte Carlo sign problem by walker annihilation in an infinite Hilbert space. They reveal the threshold for the occurrence of the first bound excited states at a value of $\alpha \approx 1.73$ for the dimensionless coupling constant. This puts the threshold into the regime of intermediate interaction strength. We calculate the spectral weight and the number density of the dressing cloud in the discretized model. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:52:39 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Taylor",
"Jamie",
""
],
[
"Čufar",
"Matija",
""
],
[
"Mitrouskas",
"David",
""
],
[
"Seiringer",
"Robert",
""
],
[
"Pahl",
"Elke",
""
],
[
"Brand",
"Joachim",
""
]
] |
2506.02441 | Zeshan Haider | Zeshan Haider, Muhammad Haroon Saeed, Muhammad Ehsan-ul-Haq Zaheer, Zeeshan Ahmed Alvi, Muhammad Ilyas, Tahira Nasreen, Muhammad Imran, Rameez Ul Islam, Manzoor Ikram | Quantum Random Number Generator (QRNG): Theoretical and Experimental Investigations | null | null | null | null | quant-ph | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Quantum Random Number Generators (QRNGs) emerged as a promising solution for generating truly random numbers. In the present article, we give an overview of QRNGs highlighting the merits and demerits of various strategies briefly. Then opting for the best-case scenario, we present the in-depth experimental explorations for building and characterizing QRNG using the homodyne detection technique to measure the quadrature amplitude of quantum vacuum fluctuations. Since entropy assessment plays a fundamental role in authenticating the true randomness, a comprehensive description of entropy and how it evaluates the quality of randomness of the source is illustrated. Our experimental setup, apart from the hardware, includes a diverse set of testing techniques including NIST statistical/entropy suites, Dieharder tests battery, and autocorrelation coefficient to verify the randomness and statistical properties of the generated random numbers. We believe that our experimental investigations provide a valuable resource for building QRNGs for a wide range of applications. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 04:55:37 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Haider",
"Zeshan",
""
],
[
"Saeed",
"Muhammad Haroon",
""
],
[
"Zaheer",
"Muhammad Ehsan-ul-Haq",
""
],
[
"Alvi",
"Zeeshan Ahmed",
""
],
[
"Ilyas",
"Muhammad",
""
],
[
"Nasreen",
"Tahira",
""
],
[
"Imran",
"Muhammad",
""
],
[
"Islam",
"Rameez Ul",
""
],
[
"Ikram",
"Manzoor",
""
]
] |
2506.02443 | Hamidou Tembine | Hamidou Tembine and Issa Bamia and Massa NDong and Bakary Coulibaly and Oumar Issiaka Traore and Moussa Traore and Moussa Sanogo and Mamadou Eric Sangare and Salif Kante and Daryl Noupa Yongueng and Hafiz Tiomoko Ali and Malik Tiomoko and Frejus Laleye and Boualem Djehiche and Wesmanegda Elisee Dipama and Idris Baba Saje and Hammid Mohammed Ibrahim and Moumini Sanogo and Marie Coursel Nininahazwe and Abdul-Latif Siita and Haine Mhlongo and Teddy Nelvy Dieu Merci Kouka and Mariam Serine Jeridi and Mutiyamuogo Parfait Mupenge and Lekoueiry Dehah and Abdoul Aziz Bio Sidi Bouko and Wilfried Franceslas Zokoue and Odette Richette Sambila and Alina RS Mbango and Mady Diagouraga and Oumarou Moussa Sanoussi and Gizachew Dessalegn and Mohamed Lamine Samoura and Bintou Laetitia Audrey Coulibaly | Breaking the Barriers of Text-Hungry and Audio-Deficient AI | 61 pages, 16 figures, 14 tables, 25 languages, 13 blockaudio per language. Presented at AI Mali, May 2025 | null | null | null | cs.SD eess.AS | http://creativecommons.org/licenses/by/4.0/ | While global linguistic diversity spans more than 7164 recognized languages, the current dominant architecture of machine intelligence remains fundamentally biased toward written text. This bias excludes over 700 million people particularly in rural and remote regions who are audio-literate. In this work, we introduce a fully textless, audio-to-audio machine intelligence framework designed to serve this underserved population, and all the people who prefer audio-efficiency. Our contributions include novel Audio-to-Audio translation architectures that bypass text entirely, including spectrogram-, scalogram-, wavelet-, and unit-based models. Central to our approach is the Multiscale Audio-Semantic Transform (MAST), a representation that encodes tonal, prosodic, speaker, and expressive features. We further integrate MAST into a fractional diffusion of mean-field-type framework powered by fractional Brownian motion. It enables the generation of high-fidelity, semantically consistent speech without reliance on textual supervision. The result is a robust and scalable system capable of learning directly from raw audio, even in languages that are unwritten or rarely digitized. This work represents a fundamental shift toward audio-native machine intelligence systems, expanding access to language technologies for communities historically left out of the current machine intelligence ecosystem. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:02:14 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Tembine",
"Hamidou",
""
],
[
"Bamia",
"Issa",
""
],
[
"NDong",
"Massa",
""
],
[
"Coulibaly",
"Bakary",
""
],
[
"Traore",
"Oumar Issiaka",
""
],
[
"Traore",
"Moussa",
""
],
[
"Sanogo",
"Moussa",
""
],
[
"Sangare",
"Mamadou Eric",
""
],
[
"Kante",
"Salif",
""
],
[
"Yongueng",
"Daryl Noupa",
""
],
[
"Ali",
"Hafiz Tiomoko",
""
],
[
"Tiomoko",
"Malik",
""
],
[
"Laleye",
"Frejus",
""
],
[
"Djehiche",
"Boualem",
""
],
[
"Dipama",
"Wesmanegda Elisee",
""
],
[
"Saje",
"Idris Baba",
""
],
[
"Ibrahim",
"Hammid Mohammed",
""
],
[
"Sanogo",
"Moumini",
""
],
[
"Nininahazwe",
"Marie Coursel",
""
],
[
"Siita",
"Abdul-Latif",
""
],
[
"Mhlongo",
"Haine",
""
],
[
"Kouka",
"Teddy Nelvy Dieu Merci",
""
],
[
"Jeridi",
"Mariam Serine",
""
],
[
"Mupenge",
"Mutiyamuogo Parfait",
""
],
[
"Dehah",
"Lekoueiry",
""
],
[
"Bouko",
"Abdoul Aziz Bio Sidi",
""
],
[
"Zokoue",
"Wilfried Franceslas",
""
],
[
"Sambila",
"Odette Richette",
""
],
[
"Mbango",
"Alina RS",
""
],
[
"Diagouraga",
"Mady",
""
],
[
"Sanoussi",
"Oumarou Moussa",
""
],
[
"Dessalegn",
"Gizachew",
""
],
[
"Samoura",
"Mohamed Lamine",
""
],
[
"Coulibaly",
"Bintou Laetitia Audrey",
""
]
] |
2506.02445 | Hong Guo | Hong Guo, Wei-Liang Qian, Bean Wang | Novel black holes with scalar hair in the Einstein-Maxwell-Scalar Theory with positive coupling | 22 pages, 7 figures | null | null | null | gr-qc | http://creativecommons.org/licenses/by-nc-sa/4.0/ | In this work, we find a new branch of hairy black hole solutions in the Einstein-Maxwell-Scalar (EMS) theory in four-dimensional asymptotically flat spacetimes. Different from spontaneous scalarization induced by tachyonic instabilities in Reissner-Nordstr\"{o}m (RN) black holes with a negative coupling parameter, these scalar-hairy black hole solutions arise when the coupling parameter is positive, where nonlinear coupling plays the dominant role, meaning that the coupling is positively correlated with the degree of deviation from the trivial state. Our numerical analysis reveals that the scalar field grows monotonically with the radial coordinate and asymptotically approaches a finite constant, exhibiting behavior that is qualitatively similar to that of the Maxwell potential. In these solutions, an increase in the charge $q$ causes the scalar-hairy solutions to deviate further from the RN state, while excessive charging drives the system back towards hairless solutions. Strengthening the coupling parameter compresses the existence domain of the scalar-hairy state, which lies entirely within the parameter region of RN black holes. Moreover, by evaluating the quasinormal modes, we show that the obtained scalar-hairy solutions are stable against linearized scalar perturbations. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:04:34 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Guo",
"Hong",
""
],
[
"Qian",
"Wei-Liang",
""
],
[
"Wang",
"Bean",
""
]
] |
2506.02446 | Jack Allsop | Jack Allsop | Isotopisms of quadratic quasigroups | null | null | null | null | math.CO math.GR | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | A quasigroup is a pair $(Q, \cdot)$ where $Q$ is a non-empty set and $\cdot$ is a binary operation on $Q$ such that for every $(u, v) \in Q^2$ there exists a unique $(x, y) \in Q^2$ such that $u \cdot x = v = y \cdot u$. Let $q$ be an odd prime power, let $\mathbb{F}_q$ denote the finite field of order $q$, and let $\mathcal{R}_q$ denote the set of non-zero squares in $\mathbb{F}_q$. Let $\{a, b\} \subseteq \mathbb{F}_q$ be such that $\{ab, (a-1)(b-1)\} \subseteq \mathcal{R}_q$. Let $\mathcal{Q}_{a, b}$ denote the quadratic quasigroup $(\mathbb{F}_q, *_{a, b})$ where $*_{a, b}$ is defined by
\[ \left\{
\begin{array}{ll}
x+a(y-x) & \text{if } y-x \in \mathcal{R}_q,\\ x+b(y-x) & \text{otherwise}.
\end{array}
\right. \] The operation table of a quadratic quasigroup is a quadratic Latin square. Recently, it has been determined exactly when two quadratic quasigroups are isomorphic and the automorphism group of any quadratic quasigroup has been determined. In this paper, we extend these results. We determine exactly when two quadratic quasigroups are isotopic and we determine the autotopism group of any quadratic quasigroup. In the process, we count the number of $2 \times 2$ subsquares in quadratic Latin squares. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:08:28 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Allsop",
"Jack",
""
]
] |
2506.02447 | Takayuki Itoh | Arisa Sugino, Takayuki Itoh | Visualization for interactively adjusting the de-bias effect of word embedding | null | null | null | null | cs.HC | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Word embedding, which converts words into numerical values, is an important natural language processing technique and widely used. One of the serious problems of word embedding is that the bias will be learned and affect the model if the dataset used for pre-training contains bias. On the other hand, indiscriminate removal of bias from word embeddings may result in the loss of information, even if the bias is undesirable to us. As a result, a risk of model performance degradation due to bias removal will be another problem. As a solution to this problem, we focus on gender bias in Japanese and propose an interactive visualization method to adjust the degree of debias for each word category. Specifically, we visualize the accuracy in a category classification task after debiasing, and allow the user to adjust the parameters based on the visualization results, so that the debiasing can be adjusted according to the user's objectives. In addition, considering a trade-off between debiasing and preventing degradation of model performance, and that different people perceive gender bias differently, we developed a mechanism to present multiple choices of debiasing configurations applying an optimization scheme. This paper presents the results of an experiment in which we removed the gender bias for word embeddings learned from the Japanese version of Wikipedia. We classified words into five categories based on a news corpus, and observed that the degree of influence of debiasing differed greatly among the categories. We then adjusted the degree of debiasing for each category based on the visualization results. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:10:42 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Sugino",
"Arisa",
""
],
[
"Itoh",
"Takayuki",
""
]
] |
2506.02448 | Baoyu Liang | Baoyu Liang, Qile Su, Shoutai Zhu, Yuchen Liang, Chao Tong | VidEvent: A Large Dataset for Understanding Dynamic Evolution of Events in Videos | null | null | null | null | cs.CV cs.AI | http://creativecommons.org/licenses/by/4.0/ | Despite the significant impact of visual events on human cognition, understanding events in videos remains a challenging task for AI due to their complex structures, semantic hierarchies, and dynamic evolution. To address this, we propose the task of video event understanding that extracts event scripts and makes predictions with these scripts from videos. To support this task, we introduce VidEvent, a large-scale dataset containing over 23,000 well-labeled events, featuring detailed event structures, broad hierarchies, and logical relations extracted from movie recap videos. The dataset was created through a meticulous annotation process, ensuring high-quality and reliable event data. We also provide comprehensive baseline models offering detailed descriptions of their architecture and performance metrics. These models serve as benchmarks for future research, facilitating comparisons and improvements. Our analysis of VidEvent and the baseline models highlights the dataset's potential to advance video event understanding and encourages the exploration of innovative algorithms and models. The dataset and related resources are publicly available at www.videvent.top. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:12:48 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Liang",
"Baoyu",
""
],
[
"Su",
"Qile",
""
],
[
"Zhu",
"Shoutai",
""
],
[
"Liang",
"Yuchen",
""
],
[
"Tong",
"Chao",
""
]
] |
2506.02449 | Bo Peng | Bo Peng, Zhiheng Wang, Heyang Gong, Chaochao Lu | IP-Dialog: Evaluating Implicit Personalization in Dialogue Systems with Synthetic Data | null | null | null | null | cs.CL cs.HC | http://creativecommons.org/licenses/by/4.0/ | In modern dialogue systems, the ability to implicitly infer user backgrounds from conversations and leverage this information for personalized assistance is crucial. However, the scarcity of high-quality data remains a fundamental challenge to evaluating and improving this capability. Traditional dataset construction methods are labor-intensive, resource-demanding, and raise privacy concerns. To address these issues, we propose a novel approach for automatic synthetic data generation and introduce the Implicit Personalized Dialogue (IP-Dialog) benchmark along with a training dataset, covering 10 tasks and 12 user attribute types. Additionally, we develop a systematic evaluation framework with four metrics to assess both attribute awareness and reasoning capabilities. We further propose five causal graphs to elucidate models' reasoning pathways during implicit personalization. Extensive experiments yield insightful observations and prove the reliability of our dataset. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:14:11 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Peng",
"Bo",
""
],
[
"Wang",
"Zhiheng",
""
],
[
"Gong",
"Heyang",
""
],
[
"Lu",
"Chaochao",
""
]
] |
2506.02450 | Dimitri Bourilkov | Dimitri Bourilkov | Muon (and Lepton) Anomalous Magnetic Moments and Limits on Their Radii | 5 pages, 3 tables | null | null | null | hep-ph hep-ex | http://creativecommons.org/licenses/by-nc-sa/4.0/ | The limits on lepton radii and contact interaction scales are used to derive limits on the deviations of lepton anomalous magnetic dipole moments (g-2)/2 from the Standard Model predictions. In the case of quadratic dependence of the deviations on the lepton mass the limits for electrons are better compared to low energy measurements, for muons somewhat weaker than the latest precision experiments, and for taus by far the best. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:15:59 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bourilkov",
"Dimitri",
""
]
] |
2506.02451 | Pratheeksha Nair | Pratheeksha Nair and Reihaneh Rabbany | Weak Supervision for Real World Graphs | null | null | null | null | cs.LG | http://creativecommons.org/licenses/by/4.0/ | Node classification in real world graphs often suffers from label scarcity and noise, especially in high stakes domains like human trafficking detection and misinformation monitoring. While direct supervision is limited, such graphs frequently contain weak signals, noisy or indirect cues, that can still inform learning. We propose WSNET, a novel weakly supervised graph contrastive learning framework that leverages these weak signals to guide robust representation learning. WSNET integrates graph structure, node features, and multiple noisy supervision sources through a contrastive objective tailored for weakly labeled data. Across three real world datasets and synthetic benchmarks with controlled noise, WSNET consistently outperforms state of the art contrastive and noisy label learning methods by up to 15% in F1 score. Our results highlight the effectiveness of contrastive learning under weak supervision and the promise of exploiting imperfect labels in graph based settings. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:16:18 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Nair",
"Pratheeksha",
""
],
[
"Rabbany",
"Reihaneh",
""
]
] |
2506.02452 | Wenshuo Chen | Wenshuo Chen, Kuimou Yu, Haozhe Jia, Kaishen Yuan, Bowen Tian, Songning Lai, Hongru Xiao, Erhang Zhang, Lei Wang, and Yutao Yue | ANT: Adaptive Neural Temporal-Aware Text-to-Motion Model | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | While diffusion models advance text-to-motion generation, their static semantic conditioning ignores temporal-frequency demands: early denoising requires structural semantics for motion foundations while later stages need localized details for text alignment. This mismatch mirrors biological morphogenesis where developmental phases demand distinct genetic programs. Inspired by epigenetic regulation governing morphological specialization, we propose **(ANT)**, an **A**daptive **N**eural **T**emporal-Aware architecture. ANT orchestrates semantic granularity through: **(i) Semantic Temporally Adaptive (STA) Module:** Automatically partitions denoising into low-frequency structural planning and high-frequency refinement via spectral analysis. **(ii) Dynamic Classifier-Free Guidance scheduling (DCFG):** Adaptively adjusts conditional to unconditional ratio enhancing efficiency while maintaining fidelity. **(iii) Temporal-semantic reweighting:** Quantitatively aligns text influence with phase requirements. Extensive experiments show that ANT can be applied to various baselines, significantly improving model performance, and achieving state-of-the-art semantic alignment on StableMoFusion. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:17:37 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Chen",
"Wenshuo",
""
],
[
"Yu",
"Kuimou",
""
],
[
"Jia",
"Haozhe",
""
],
[
"Yuan",
"Kaishen",
""
],
[
"Tian",
"Bowen",
""
],
[
"Lai",
"Songning",
""
],
[
"Xiao",
"Hongru",
""
],
[
"Zhang",
"Erhang",
""
],
[
"Wang",
"Lei",
""
],
[
"Yue",
"Yutao",
""
]
] |
2506.02453 | Kunyu Wang | Kunyu Wang, Xueyang Fu, Yunfei Bao, Chengjie Ge, Chengzhi Cao, Wei Zhai, Zheng-Jun Zha | PAID: Pairwise Angular-Invariant Decomposition for Continual Test-Time Adaptation | null | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-nd/4.0/ | Continual Test-Time Adaptation (CTTA) aims to online adapt a pre-trained model to changing environments during inference. Most existing methods focus on exploiting target data, while overlooking another crucial source of information, the pre-trained weights, which encode underutilized domain-invariant priors. This paper takes the geometric attributes of pre-trained weights as a starting point, systematically analyzing three key components: magnitude, absolute angle, and pairwise angular structure. We find that the pairwise angular structure remains stable across diverse corrupted domains and encodes domain-invariant semantic information, suggesting it should be preserved during adaptation. Based on this insight, we propose PAID (Pairwise Angular-Invariant Decomposition), a prior-driven CTTA method that decomposes weight into magnitude and direction, and introduces a learnable orthogonal matrix via Householder reflections to globally rotate direction while preserving the pairwise angular structure. During adaptation, only the magnitudes and the orthogonal matrices are updated. PAID achieves consistent improvements over recent SOTA methods on four widely used CTTA benchmarks, demonstrating that preserving pairwise angular structure offers a simple yet effective principle for CTTA. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:18:15 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Wang",
"Kunyu",
""
],
[
"Fu",
"Xueyang",
""
],
[
"Bao",
"Yunfei",
""
],
[
"Ge",
"Chengjie",
""
],
[
"Cao",
"Chengzhi",
""
],
[
"Zhai",
"Wei",
""
],
[
"Zha",
"Zheng-Jun",
""
]
] |
2506.02454 | Zhaorui Yang | Zhaorui Yang, Bo Pan, Han Wang, Yiyao Wang, Xingyu Liu, Minfeng Zhu, Bo Zhang, Wei Chen | Multimodal DeepResearcher: Generating Text-Chart Interleaved Reports From Scratch with Agentic Framework | 47 pages | null | null | null | cs.CL cs.AI | http://creativecommons.org/licenses/by/4.0/ | Visualizations play a crucial part in effective communication of concepts and information. Recent advances in reasoning and retrieval augmented generation have enabled Large Language Models (LLMs) to perform deep research and generate comprehensive reports. Despite its progress, existing deep research frameworks primarily focus on generating text-only content, leaving the automated generation of interleaved texts and visualizations underexplored. This novel task poses key challenges in designing informative visualizations and effectively integrating them with text reports. To address these challenges, we propose Formal Description of Visualization (FDV), a structured textual representation of charts that enables LLMs to learn from and generate diverse, high-quality visualizations. Building on this representation, we introduce Multimodal DeepResearcher, an agentic framework that decomposes the task into four stages: (1) researching, (2) exemplar report textualization, (3) planning, and (4) multimodal report generation. For the evaluation of generated multimodal reports, we develop MultimodalReportBench, which contains 100 diverse topics served as inputs along with 5 dedicated metrics. Extensive experiments across models and evaluation methods demonstrate the effectiveness of Multimodal DeepResearcher. Notably, utilizing the same Claude 3.7 Sonnet model, Multimodal DeepResearcher achieves an 82\% overall win rate over the baseline method. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:18:19 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Yang",
"Zhaorui",
""
],
[
"Pan",
"Bo",
""
],
[
"Wang",
"Han",
""
],
[
"Wang",
"Yiyao",
""
],
[
"Liu",
"Xingyu",
""
],
[
"Zhu",
"Minfeng",
""
],
[
"Zhang",
"Bo",
""
],
[
"Chen",
"Wei",
""
]
] |
2506.02455 | Ian Wanless | Jack Allsop and Ian M. Wanless | Perfect $1$-factorisations of $K_{11,11}$ | null | null | null | null | math.CO | http://creativecommons.org/licenses/by-nc-nd/4.0/ | A perfect $1$-factorisation of a graph is a decomposition of that graph into $1$-factors such that the union of any two $1$-factors is a Hamiltonian cycle. A Latin square of order $n$ is row-Hamiltonian if for every pair $(r,s)$ of distinct rows, the permutation mapping $r$ to $s$ has a single cycle of length $n$. We report the results of a computer enumeration of the perfect $1$-factorisations of the complete bipartite graph $K_{11,11}$. This also allows us to find all row-Hamiltonian Latin squares of order $11$. Finally, we plug a gap in the literature regarding how many row-Hamiltonian Latin squares are associated with the classical families of perfect $1$-factorisations of complete graphs. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:20:54 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Allsop",
"Jack",
""
],
[
"Wanless",
"Ian M.",
""
]
] |
2506.02456 | Tr\'i Cao | Tri Cao, Bennett Lim, Yue Liu, Yuan Sui, Yuexin Li, Shumin Deng, Lin Lu, Nay Oo, Shuicheng Yan, Bryan Hooi | VPI-Bench: Visual Prompt Injection Attacks for Computer-Use Agents | Under Review | null | null | null | cs.AI cs.CR | http://creativecommons.org/licenses/by/4.0/ | Computer-Use Agents (CUAs) with full system access enable powerful task automation but pose significant security and privacy risks due to their ability to manipulate files, access user data, and execute arbitrary commands. While prior work has focused on browser-based agents and HTML-level attacks, the vulnerabilities of CUAs remain underexplored. In this paper, we investigate Visual Prompt Injection (VPI) attacks, where malicious instructions are visually embedded within rendered user interfaces, and examine their impact on both CUAs and Browser-Use Agents (BUAs). We propose VPI-Bench, a benchmark of 306 test cases across five widely used platforms, to evaluate agent robustness under VPI threats. Each test case is a variant of a web platform, designed to be interactive, deployed in a realistic environment, and containing a visually embedded malicious prompt. Our empirical study shows that current CUAs and BUAs can be deceived at rates of up to 51% and 100%, respectively, on certain platforms. The experimental results also indicate that system prompt defenses offer only limited improvements. These findings highlight the need for robust, context-aware defenses to ensure the safe deployment of multimodal AI agents in real-world environments. The code and dataset are available at: https://github.com/cua-framework/agents | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:21:50 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Cao",
"Tri",
""
],
[
"Lim",
"Bennett",
""
],
[
"Liu",
"Yue",
""
],
[
"Sui",
"Yuan",
""
],
[
"Li",
"Yuexin",
""
],
[
"Deng",
"Shumin",
""
],
[
"Lu",
"Lin",
""
],
[
"Oo",
"Nay",
""
],
[
"Yan",
"Shuicheng",
""
],
[
"Hooi",
"Bryan",
""
]
] |
2506.02457 | Yixuan Hou | Yixuan Hou, Heyang Liu, Yuhao Wang, Ziyang Cheng, Ronghua Wu, Qunshan Gu, Yanfeng Wang, Yu Wang | SOVA-Bench: Benchmarking the Speech Conversation Ability for LLM-based Voice Assistant | null | null | null | null | cs.SD eess.AS | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Thanks to the steady progress of large language models (LLMs), speech encoding algorithms and vocoder structure, recent advancements have enabled generating speech response directly from a user instruction. However, benchmarking the generated speech quality has been a neglected but critical issue, considering the shift from the pursuit of semantic accuracy to vivid and spontaneous speech flow. Previous evaluation focused on the speech-understanding ability, lacking a quantification of acoustic quality. In this paper, we propose Speech cOnversational Voice Assistant Benchmark (SOVA-Bench), providing a comprehension comparison of the general knowledge, speech recognition and understanding, along with both semantic and acoustic generative ability between available speech LLMs. To the best of our knowledge, SOVA-Bench is one of the most systematic evaluation frameworks for speech LLMs, inspiring the direction of voice interaction systems. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:21:51 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Hou",
"Yixuan",
""
],
[
"Liu",
"Heyang",
""
],
[
"Wang",
"Yuhao",
""
],
[
"Cheng",
"Ziyang",
""
],
[
"Wu",
"Ronghua",
""
],
[
"Gu",
"Qunshan",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Wang",
"Yu",
""
]
] |
2506.02458 | Trinh Van Chien | Nguyen Chi Long, Trinh Van Chien, Ta Hai Tung, Van Son Nguyen, Trong-Minh Hoang, Nguyen Ngoc Hai Dang | A Novel Deep Reinforcement Learning Method for Computation Offloading in Multi-User Mobile Edge Computing with Decentralization | 6 pages, 5 figures, and 1 table. Published by IEEE at ATC2024 | null | null | null | cs.IT cs.LG math.IT | http://arxiv.org/licenses/nonexclusive-distrib/1.0/ | Mobile edge computing (MEC) allows appliances to offload workloads to neighboring MEC servers that have the potential for computation-intensive tasks with limited computational capabilities. This paper studied how deep reinforcement learning (DRL) algorithms are used in an MEC system to find feasible decentralized dynamic computation offloading strategies, which leads to the construction of an extensible MEC system that operates effectively with finite feedback. Even though the Deep Deterministic Policy Gradient (DDPG) algorithm, subject to their knowledge of the MEC system, can be used to allocate powers of both computation offloading and local execution, to learn a computation offloading policy for each user independently, we realized that this solution still has some inherent weaknesses. Hence, we introduced a new approach for this problem based on the Twin Delayed DDPG algorithm, which enables us to overcome this proneness and investigate cases where mobile users are portable. Numerical results showed that individual users can autonomously learn adequate policies through the proposed approach. Besides, the performance of the suggested solution exceeded the conventional DDPG-based power control strategy. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:22:02 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Long",
"Nguyen Chi",
""
],
[
"Van Chien",
"Trinh",
""
],
[
"Tung",
"Ta Hai",
""
],
[
"Nguyen",
"Van Son",
""
],
[
"Hoang",
"Trong-Minh",
""
],
[
"Dang",
"Nguyen Ngoc Hai",
""
]
] |
2506.02459 | Martin Juan Jos\'e Bucher | Martin JJ. Bucher, Iro Armeni | ReSpace: Text-Driven 3D Scene Synthesis and Editing with Preference Alignment | 20 pages, 17 figures (incl. appendix) | null | null | null | cs.CV | http://creativecommons.org/licenses/by-nc-sa/4.0/ | Scene synthesis and editing has emerged as a promising direction in computer graphics. Current trained approaches for 3D indoor scenes either oversimplify object semantics through one-hot class encodings (e.g., 'chair' or 'table'), require masked diffusion for editing, ignore room boundaries, or rely on floor plan renderings that fail to capture complex layouts. In contrast, LLM-based methods enable richer semantics via natural language (e.g., 'modern studio with light wood furniture') but do not support editing, remain limited to rectangular layouts or rely on weak spatial reasoning from implicit world models. We introduce ReSpace, a generative framework for text-driven 3D indoor scene synthesis and editing using autoregressive language models. Our approach features a compact structured scene representation with explicit room boundaries that frames scene editing as a next-token prediction task. We leverage a dual-stage training approach combining supervised fine-tuning and preference alignment, enabling a specially trained language model for object addition that accounts for user instructions, spatial geometry, object semantics, and scene-level composition. For scene editing, we employ a zero-shot LLM to handle object removal and prompts for addition. We further introduce a novel voxelization-based evaluation that captures fine-grained geometry beyond 3D bounding boxes. Experimental results surpass state-of-the-art on object addition while maintaining competitive results on full scene synthesis. | [
{
"version": "v1",
"created": "Tue, 03 Jun 2025 05:22:04 GMT"
}
] | 2025-06-04T00:00:00 | [
[
"Bucher",
"Martin JJ.",
""
],
[
"Armeni",
"Iro",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.