Hugging Face's logo

TIGER-Lab
/
VL-Rethinker-7B

Image-Text-to-Text
Transformers
Safetensors
English
qwen2_5_vl
multimodal
conversational
text-generation-inference
Model card Files Community
1
JasperHaozhe commited on
Commit
f720632
·
verified ·
1 Parent(s): 52ef36e

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -1
README.md CHANGED
@@ -13,9 +13,11 @@ library_name: transformers
13
 
14
  # VL-Rethinker-7B
15
 
 
 
16
  **VL-Rethinker-7B** achieves SoTA results on various multimodal reasoning benchmarks.
17
 
18
- It is trained using the **GRPO-SSR and Forced Rethinking** techniques, using meticulously curated **training data**: [ViRL39K](https://huggingface.co/datasets/TIGER-Lab/ViRL39K).
19
 
20
  For details of our approach and performance comparison, please see our [paper](https://github.com/TIGER-AI-Lab/VL-Rethinker/blob/main/paper.pdf).
21
 
 
13
 
14
  # VL-Rethinker-7B
15
 
16
+ **🚀 News:** <u>We release our meticulously curated collection of RL training queries for multimodal reasoning: [ViRL39K](https://huggingface.co/datasets/TIGER-Lab/ViRL39K).</u>
17
+
18
  **VL-Rethinker-7B** achieves SoTA results on various multimodal reasoning benchmarks.
19
 
20
+ It is trained using the **GRPO-SSR and Forced Rethinking** techniques, using meticulously curated [ViRL39K](https://huggingface.co/datasets/TIGER-Lab/ViRL39K).
21
 
22
  For details of our approach and performance comparison, please see our [paper](https://github.com/TIGER-AI-Lab/VL-Rethinker/blob/main/paper.pdf).
23