Antimage01 commited on
Commit
a4ca6f8
·
verified ·
1 Parent(s): 405f887

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +48 -3
README.md CHANGED
@@ -1,3 +1,48 @@
1
- ---
2
- license: apache-2.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ datasets:
3
+ - URSA-MATH/MMathCoT-1M
4
+ language:
5
+ - en
6
+ - zh
7
+ license: apache-2.0
8
+ library_name: transformers
9
+ pipeline_tag: image-text-to-text
10
+ ---
11
+
12
+
13
+ # URSA-8B-PS-GRPO
14
+
15
+ URSA-8B-PS-GRPO employs process-supervision grpo which proposed in our [paper](https://arxiv.org/pdf/2501.04686).
16
+
17
+ # Installation
18
+
19
+ ```python
20
+ from huggingface_hub import snapshot_download
21
+
22
+ repo_id = "URSA-MATH/URSA-8B-PS-GRPO"
23
+ local_dir = YOUR_LOCAL_PATH
24
+
25
+ snapshot_path = snapshot_download(
26
+ repo_id=repo_id,
27
+ local_dir=local_dir,
28
+ revision="main",
29
+ cache_dir=None,
30
+ )
31
+ ```
32
+ # Inference
33
+ We have adapted vLLM for URSA-8B. Please refer to the [GitHub](https://github.com/URSA-MATH/URSA-MATH) repository for quick inference implementation.
34
+
35
+ Besides, we have adapted evaluation on [VLMEvalKit](https://github.com/open-compass/VLMEvalKit)!
36
+
37
+ # Citation
38
+
39
+ If you find our paper, model, or data helpful, please give this repo a star 🌟 and cite our article ✏️.
40
+ ```
41
+ @article{luo2025ursa,
42
+ title={URSA: Understanding and Verifying Chain-of-thought Reasoning in Multimodal Mathematics},
43
+ author={Luo, Ruilin and Zheng, Zhuofan and Wang, Yifan and Yu, Yiyao and Ni, Xinzhe and Lin, Zicheng and Zeng, Jin and Yang, Yujiu},
44
+ journal={arXiv preprint arXiv:2501.04686},
45
+ year={2025}
46
+ }
47
+ ```
48
+ ```