dMLLM-TTS: Self-Verified and Efficient Test-Time Scaling for Diffusion Multi-Modal Large Language Models
Abstract
dMLLM-TTS improves the efficiency and quality of diffusion multi-modal large language models through a hierarchical search algorithm and self-verified feedback mechanism.
Diffusion Multi-modal Large Language Models (dMLLMs) have recently emerged as a novel architecture unifying image generation and understanding. However, developing effective and efficient Test-Time Scaling (TTS) methods to unlock their full generative potential remains an underexplored challenge. To address this, we propose dMLLM-TTS, a novel framework operating on two complementary scaling axes: (1) trajectory exploration scaling to enhance the diversity of generated hypotheses, and (2) iterative refinement scaling for stable generation. Conventional TTS approaches typically perform linear search across these two dimensions, incurring substantial computational costs of O(NT) and requiring an external verifier for best-of-N selection. To overcome these limitations, we propose two innovations. First, we design an efficient hierarchical search algorithm with O(N+T) complexity that adaptively expands and prunes sampling trajectories. Second, we introduce a self-verified feedback mechanism that leverages the dMLLMs' intrinsic image understanding capabilities to assess text-image alignment, eliminating the need for external verifier. Extensive experiments on the GenEval benchmark across three representative dMLLMs (e.g., Lumina-DiMOO, MMaDA, Muddit) show that our framework substantially improves generation quality while achieving up to 6x greater efficiency than linear search. Project page: https://github.com/Alpha-VLLM/Lumina-DiMOO.
Community
Pioneering TTS Framework: we establish the first test-time scaling framework for dMLLMs that integrates scaling strategy, verification, and search algorithm.
Novel Verifier: we introduce a self-verification mechanism that leverages the model’s inherent image understanding to internally evaluate generative outcomes, eliminating external verifiers.
Efficient Search Algorithm: we present the Hierarchical Trajectory Search, which optimizes efficiency, achieving
O(N +T) complexity, outperforming conventional linear search baseline with O(NT) complexity.Superior Performance: the proposed TTS framework elevates dMLLMs to match state-of-the-art generation
models, significantly boosting image quality.
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
