Add DeepResearch Bench application with LFS support

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+*.jsonl filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,11 @@
+__pycache__/
+*.py[cod]
+*$py.class
+.env
+.venv
+env/
+venv/
+ENV/
+.DS_Store
+*.log
+data/data_viewer.jsonl 

Dockerfile

@@ -0,0 +1,11 @@
+FROM python:3.10-slim
+
+WORKDIR /code
+
+COPY . /code/
+
+RUN pip install --no-cache-dir --upgrade pip && \
+    pip install --no-cache-dir -r requirements.txt
+
+# 默认运行命令
+CMD ["python", "app.py"] 

README.md

@@ -1,12 +1,25 @@
----
-title: DeepResearch Bench
-emoji: 🏆
-colorFrom: green
-colorTo: red
-sdk: gradio
-sdk_version: 5.31.0
-app_file: app.py
-pinned: false
----
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# DeepResearch Bench
+
+**DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents**
+
+This application showcases comprehensive evaluation results for Deep Research Agents. The app includes:
+
+- 🏆 **Leaderboard** - View overall performance metrics across all evaluated models
+- 🔍 **Data Viewer** - Explore detailed results for individual research tasks
+- 📊 **Side-by-Side Comparison** - Compare different models' responses to the same research questions
+
+Visit our [project website](https://deepresearch-bench.github.io) for more information.
+
+## Citation
+```bibtex
+@article{du2025deepresearch,
+  author    = {Mingxuan Du and Benfeng Xu and Chiwei Zhu and Xiaorui Wang and Zhendong Mao},
+  title     = {DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents},
+  journal   = {arXiv preprint},
+  year      = {2025},
+}
+```
+
+## Hugging Face Space Details
+- SDK: Gradio
+- SDK Version: 3.50.0 

app.py

@@ -0,0 +1,16 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+DeepResearch Bench HF Space 入口文件
+"""
+
+from __future__ import annotations
+from create_leaderboard import demo
+
+# 在Hugging Face Space中运行
+if __name__ == "__main__":
+    demo.launch(
+        server_name="0.0.0.0",  # 必须这样设置以允许外部访问
+        share=False,  # HF Space 自己有分享功能，无需额外分享
+        show_api=False,  # 隐藏API文档页面
+    ) 

create_leaderboard.py

@@ -0,0 +1,91 @@
+#!/usr/bin/env python3
+# -*- coding: utf-8 -*-
+"""
+Gradio UI – v2.1  (Leaderboard · Data Viewer · Prompt-to-Leaderboard)
+"""
+
+from __future__ import annotations
+from pathlib import Path
+import gradio as gr
+
+# ---- Tab 组件 ----
+from tabs.leaderboard_tab import create_leaderboard_tab
+from tabs.data_viewer_tab import create_data_viewer_tab
+from tabs.data_viewer_side_by_side_tab import create_data_viewer_side_by_side_tab
+
+# ---------------------------------------------------------------------------
+# UI
+# ---------------------------------------------------------------------------
+
+with gr.Blocks(title="DeepResearch Bench") as demo:
+
+    # ========= 全局 CSS（仅作用于自定义标题 & 简介） =========
+    gr.HTML("""
+    <style>
+      .title-block{
+        /* 渐变文字效果 - 改进版 */
+        background: linear-gradient(to right, #009CFF, #823AFF);
+        background: -webkit-linear-gradient(to right, #009CFF, #823AFF);
+        background: -moz-linear-gradient(to right, #009CFF, #823AFF);
+        -webkit-background-clip: text;
+        -webkit-text-fill-color: transparent;
+        background-clip: text;
+        color: transparent;
+        
+        text-align: center;
+        font-size: 2.1rem;
+        font-weight: 700;
+        margin: 0 0 1rem 0;
+        padding-bottom: 0.2rem;
+        display: inline-block; /* 重要：确保渐变效果正常 */
+        width: 100%; /* 确保居中对齐 */
+      }
+      .intro-block{
+        text-align:center;
+        margin-bottom:1.25rem;
+        line-height:2;
+      }
+      .intro-block a{
+        color:#0a58ca;
+        text-decoration:none;
+        margin:0 .3rem;
+      }
+      .intro-block a:hover{ text-decoration:underline; }
+    </style>
+    """)
+
+    # ========= 顶部标题 & 简介（不使用 Markdown 标题语法） =========
+    gr.HTML("""
+    <div class="title-block">
+      DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents
+    </div>
+
+    <div class="intro-block">
+      The research aims to comprehensively evaluate the capabilities of Deep Research Agents.<br>
+      <a href="#">Code</a> |
+      <a href="#">Website</a> |
+      <a href="#">Paper</a> |
+      <a href="#">Eval Dataset</a> |
+      Total models: 16 | Last Update: 28 May 2025
+    </div>
+    """)
+
+    # ========= 主 Tabs =========
+    with gr.Tabs():
+        create_leaderboard_tab()           # 🏆 Leaderboard
+        create_data_viewer_tab()           # 🔍 Data Viewer
+        create_data_viewer_side_by_side_tab()
+
+        with gr.Tab("💬Prompt-to-Leaderboard"):
+            gr.Markdown(
+                """
+🚧 **Prompt-to-Leaderboard** module not implemented yet.  
+Planned: inspect how individual prompts affect overall model ranking.
+"""
+            )
+
+# ---------------------------------------------------------------------------
+# Entrypoint
+# ---------------------------------------------------------------------------
+if __name__ == "__main__":
+    demo.launch()

data/data_viewer.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a7ab11f250f4ffd6bf9c74ff8dc1e68f86d7abbf4f6319164bb476177ad7bf6e
+size 28044256

data/leaderboard.csv

@@ -0,0 +1,17 @@
+model,overall_score,comprehensiveness,insight,instruction_following,readability,citation_accuracy,effective_citations
+gemini-2.5-pro-deepresearch,48.88,48.53,48.50,49.18,49.44,81.44,111.21
+openai-deepresearch,46.98,46.87,45.25,49.27,47.14,77.96,40.79
+perplexity-Research,42.25,40.69,39.39,46.40,44.28,90.24,31.26
+claude-3-7-sonnet-with-search,40.67,38.99,37.66,45.77,41.46,93.68,32.48
+grok-deeper-search,40.24,37.97,35.37,46.30,44.05,83.59,8.15
+perplexity-sonar-reasoning-pro,40.22,37.38,36.11,45.66,44.74,39.36,8.35
+perplexity-sonar-reasoning,40.18,37.14,36.73,45.15,44.35,48.67,11.34
+perplexity-sonar-pro,38.93,36.38,34.26,44.70,43.35,78.66,14.74
+gemini-2.5-pro-with-grounding,35.12,34.06,29.79,41.67,37.16,81.81,32.88
+gpt-4o-search-preview,35.10,31.99,27.57,43.17,41.23,88.41,4.79
+perplexity-sonar,34.54,30.95,27.51,42.33,41.60,74.42,8.67
+gpt-4.1-with-search,33.46,29.42,25.38,42.33,40.77,87.83,4.42
+gemini-2.5-flash-preview-04-17,32.39,31.63,26.73,38.82,34.48,81.92,31.08
+gpt-4o-mini-search-preview,31.55,27.38,22.64,40.67,39.91,84.98,4.95
+gpt-4.1-mini-with-search,30.26,26.05,20.75,39.65,39.33,84.58,4.35
+claude-3-5-sonnet-with-search,28.48,24.82,22.82,35.12,35.08,94.04,9.78

data/raw_data/claude-3-5-sonnet-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8769fc2e0cf4f059da6e34839f9df09a6fdab9e2872faa467eafa1aa42316a69
+size 505860

data/raw_data/claude-3-7-sonnet-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc16f997d3ecd09bccf6d9e756d9ad36d2834d2ed0827b8f39579b6321b98837
+size 2281964

data/raw_data/gemini-2.5-flash-with-grounding.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:908295fc145ac2f8833396b56eac8913726d93288ad6b93a9c01a69cbdbbf78a
+size 1016172

data/raw_data/gemini-2.5-pro-deepresearch.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33c5d28e76595f22fae1b0fbbe2700958bfe707dafe53f7c5842d3067ccfddef
+size 8523353

data/raw_data/gemini-2.5-pro-with-grounding.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b409b1031cff2876fd20cd8e9fc95501f2a95ad0154d3634b6538c165373447
+size 1050267

data/raw_data/gpt-4.1-mini-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf809806c294364bb45cb337355d360b0e5e023c8e4ffdbf9557880a02137bab
+size 463012

data/raw_data/gpt-4.1-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e0228624b09e9d6c25c72156f4dd7f5702e3adcdd71a1f309094c2913eb50639
+size 492406

data/raw_data/gpt-4o-mini-search-preview.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd49e75b1e7eb6ff40cd4c030032459d727987dd298863b488b9657ae18815a1
+size 541532

data/raw_data/gpt-4o-search-preview.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfb9de873345d6789197013f0cd60fb2d888957bc123447f2a8486e81c296f04
+size 565183

data/raw_data/grok-deeper-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea6428dcf2e729d84f019c302fb3862a85cefbea08282b5ffcc5c400306ab077
+size 1149933

data/raw_data/openai-deepresearch.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:77d31b8ea1abd9aa8e924303451dc6a0f334f2e9d4d61ec71847c4db004ac62a
+size 6903938

data/raw_data/perplexity-Research.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f27cb31cbab84f60efc3286592e84690fd117355dd84f9e4a9299108245c2a5
+size 1747979

data/raw_data/perplexity-sonar-pro.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d577c0a208b35eb2c0454c00c70b12759cd8a1687f730f2133d8f392c1831ee
+size 750234

data/raw_data/perplexity-sonar-reasoning-pro.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e08c6c4094bbf0aa1749e7b1a45e856a6635b2df6afdf0de8eeafea99e7477fc
+size 495156

data/raw_data/perplexity-sonar-reasoning.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5aecbee30882b3ccd2d65470526fe48c7c016869f00593933a35e7096fe4fb74
+size 659883

data/raw_data/perplexity-sonar.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc0ef26282e404b700d56e158644f44228c49a3d5126fa12c8068e053444131e
+size 574856

data/raw_results/claude-3-5-sonnet-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0c47d1bab126886420bd53bb41a8905cdfb97f105711bcc2f5a27e3d53652ea
+size 1992421

data/raw_results/claude-3-7-sonnet-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b3a6bf74400c89d24fa47853ab034ed3696ee0694c2d190ba83c3f5dcd8a0ef
+size 2002379

data/raw_results/gemini-2.5-flash-with-grounding.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43b9f71819babb5c00f65f0dd71d707323fb803c585bd74976f49cdc34ab80aa
+size 1951481

data/raw_results/gemini-2.5-pro-deepresearch.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac2fc53c99697e3276c98d735ed630df6fa49d2972c70a5409adc1958ecaa7b7
+size 1937730

data/raw_results/gemini-2.5-pro-with-grounding.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5e911a18cf8b8a8207eb45584ac650e4640f79db7352055ca5e92356de37f911
+size 1944815

data/raw_results/gpt-4.1-mini-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:948a403d12bcf6b0e3ce6664f83afeb95413684ab0b7912003ed756a4df15c5e
+size 1992345

data/raw_results/gpt-4.1-with-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:908a5989af337e381bf2bce6795438edd21966f313b5194f532feb1f47e5b812
+size 2090582

data/raw_results/gpt-4o-mini-search-preview.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4277a9a91fcdaaeff1afe948c1088095d5f01092404fcd1a62407b7a58b7906e
+size 2074673

data/raw_results/gpt-4o-search-preview.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7adcd70d49d3b5dd6050201aa4fcd31f51288945f4a23de14432a301cbf295a7
+size 2063854

data/raw_results/grok-deeper-search.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b19fb7ec93872317eae94abeb02ed9c19912057acfa82600167ca853b750f476
+size 1968989

data/raw_results/openai-deepresearch.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae45c25f5b5c56a772331543e4eefe7c80e63f33b441dfe83cb4a5c830c88a35
+size 2007501

data/raw_results/perplexity-Research.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7715271d17cc344873653464ae3fef884e0f3c6bec89deee347ed7a0651beb9
+size 2030483

data/raw_results/perplexity-sonar-pro.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a453f5b29492f684f53364121e7c79eeb81aee2737a383e2748830a4e4453afb
+size 1975770

data/raw_results/perplexity-sonar-reasoning-pro.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:669a4a14232c63c716de766af7be050f8712f74a6d5437cc8fa637ded39f3c40
+size 1957092

data/raw_results/perplexity-sonar-reasoning.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bceb5637a9d0092af5ddcca49557a4f8f3604be9ebb430be32e820fa4d6723b3
+size 1951258

data/raw_results/perplexity-sonar.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36ecd1540447863f66bfe1a43905070f9c9b0d40de803348c3450a396df3d8fc
+size 2016838

requirements.txt

@@ -0,0 +1,6 @@
+gradio>=3.50.0
+pandas
+numpy
+plotly
+pathlib
+requests