Upload README.md with huggingface_hub

README.md

@@ -30,13 +30,13 @@ datasets:
 
 ## BioReason-Pro RL
 
-Reinforcement learning (GRPO) optimized checkpoint of BioReason-Pro — a multimodal reasoning LLM for protein function prediction. This model builds on the SFT checkpoint and is further optimized through group relative policy optimization to improve reasoning quality and GO term prediction accuracy.
+Reinforcement learning (GRPO) optimized checkpoint of BioReason-Pro, a multimodal reasoning LLM for protein function prediction. This model builds on the SFT checkpoint and is further optimized through group relative policy optimization to improve reasoning quality and GO term prediction accuracy.
 
-**Training data:** 9,197 proteins, available at [wanglab/bioreason-pro-rl-reasoning-data](https://huggingface.co/datasets/wanglab/bioreason-pro-rl-reasoning-data).
+**Training data:** [wanglab/bioreason-pro-rl-reasoning-data](https://huggingface.co/datasets/wanglab/bioreason-pro-rl-reasoning-data)
 
 See also:
-- [BioReason-Pro SFT](https://huggingface.co/wanglab/bioreason-pro-sft) — supervised fine-tuned checkpoint
-- [GO-GPT](https://huggingface.co/wanglab/gogpt) — autoregressive GO term predictor
+- [BioReason-Pro SFT](https://huggingface.co/wanglab/bioreason-pro-sft) - supervised fine-tuned checkpoint
+- [GO-GPT](https://huggingface.co/wanglab/gogpt) - autoregressive GO term predictor
 
 
 ## Citation