Popular repositories Loading

1. P2T P2T Public

   P2T: Unlocking Token Rewards via Training-Free Reward Attribution (CVPR 2026) - official code

   Python 79 7

2. AC-OPD AC-OPD Public

   Official Code of ACL-2026 Paper "TRAC: Teacher-Guided Token Reward with Adaptive Calibration for Robust Policy Optimization"

   Python 42 2

3. Dataset-Tool-Segmentation Dataset-Tool-Segmentation Public

   An Auto-Process Tool to Download and Reorganize some Common Segmentation Datasets

   Python 22 2

4. SR-TTRL SR-TTRL Public

   Official Code of ICML-2026 Paper "Beyond Majority Voting: Self-Reflective Test-Time Reinforcement Learning for LLM Reasoning"

   Python 20 2

5. SaCo-Loss SaCo-Loss Public

   Official Code of CVPR-2024 Paper "SaCo Loss: Sample-wise Affinity Consistency for Vision-Language Pre-training"

   Python 18 3

6. DLoFT DLoFT Public

   Official Code of NeurIPS-2025 Paper "DLoFT: Gradient-Decoupled Fine-Tuning for Generalizable Long Chain-of-Thought Reasoning"

   Python 14 2