@@ -7,18 +7,26 @@ A unified evaluation toolkit and leaderboard for rigorously assessing the scient
77<hr style =" width :100% ;margin :16px 0 ;border :0 ;border-top :0.1px solid #d0d7de ;" />
88
99<p align =" center " >
10- <a href =" https://opencompass.org.cn/Intern-Discovery-Eval " style =" text-decoration :none ;border-bottom :0 ;" >
11- <img src="https://img.shields.io/badge/Website-SciEval-b8dcff?style=for-the-badge&logo=google-chrome&logoColor=white" />
12- </a >  ;  ;  ;
13- <a href =" https://huggingface.co/spaces/InternScience/SciEval-Leaderboard " style =" text-decoration :none ;border-bottom :0 ;" >
14- <img src="https://img.shields.io/badge/LEADERBOARD-Scieval-f6e58d?style=for-the-badge&logo=huggingface" />
15- </a >  ;  ;  ;
16- <a href =" https://github.com/InternScience/SciEvalKit/blob/main/docs/SciEvalKit.pdf " style =" text-decoration :none ;border-bottom :0 ;" >
17- <img src="https://img.shields.io/badge/REPORT-Technical-f4c2d7?style=for-the-badge" />
18- </a >  ;  ;  ;
19- <a href =" https://github.com/InternScience/SciEvalKit " style =" text-decoration :none ;border-bottom :0 ;" >
20- <img src="https://img.shields.io/badge/GitHub-Repository-c7b9e2?style=for-the-badge&logo=github&logoColor=white" />
21- </a >
10+ <span style =" display :inline-block ;vertical-align :middle ;" >
11+ <a href="https://opencompass.org.cn/Intern-Discovery-Eval" style="text-decoration:none;border-bottom:0;">
12+ <img src="https://img.shields.io/badge/Website-SciEval-b8dcff?style=for-the-badge&logo=google-chrome&logoColor=white" style="display:block;" />
13+ </a>
14+ </span >  ;  ;  ;
15+ <span style =" display :inline-block ;vertical-align :middle ;" >
16+ <a href="https://huggingface.co/spaces/InternScience/SciEval-Leaderboard" style="text-decoration:none;border-bottom:0;">
17+ <img src="https://img.shields.io/badge/LEADERBOARD-Scieval-f6e58d?style=for-the-badge&logo=huggingface" style="display:block;" />
18+ </a>
19+ </span >  ;  ;  ;
20+ <span style =" display :inline-block ;vertical-align :middle ;" >
21+ <a href="https://github.com/InternScience/SciEvalKit/blob/main/docs/SciEvalKit.pdf" style="text-decoration:none;border-bottom:0;">
22+ <img src="https://img.shields.io/badge/REPORT-Technical-f4c2d7?style=for-the-badge" style="display:block;" />
23+ </a>
24+ </span >  ;  ;  ;
25+ <span style =" display :inline-block ;vertical-align :middle ;" >
26+ <a href="https://github.com/InternScience/SciEvalKit" style="text-decoration:none;border-bottom:0;">
27+ <img src="https://img.shields.io/badge/GitHub-Repository-c7b9e2?style=for-the-badge&logo=github&logoColor=white" style="display:block;" />
28+ </a>
29+ </span >
2230</p >
2331
2432<p align =" center " >
@@ -67,6 +75,16 @@ Its design is shaped by following core ideas:
6775| ** Multimodal & Executable Scoring** | Supports text, code, and image inputs; integrates code tasks and LLM-judge fallback for open-ended answers. |
6876| ** Reproducible & Extensible** | Clear dataset and model registries, minimised hard-coding and modular evaluators make new tasks or checkpoints easy to plug in. |
6977
78+ <div align =" center " >
79+ <img src =" assets/framework.png " alt =" SciEval framework overview " width =" 65% " >
80+ </div >
81+
82+ <p align =" center " >
83+ <em >
84+ An overview of the SciEval framework, illustrating how heterogeneous scientific datasets, unified prompt construction, model inference, and capability-oriented evaluators are integrated into a single reproducible evaluation pipeline.
85+ </em >
86+ </p >
87+
7088
7189## <img src =" assets/icon/news.png " alt =" news " height =" 28 " style =" vertical-align :middle ;" />  ; News
7290* ** [ 2025‑12‑12] · 📰 Evaluation Published on OpenCompass**
0 commit comments