Pinned Loading

1. SAGE SAGE Public

   Self-hinting RL increases the usage rate of hard prompts, and improves LLM's performance.

   Python 19 3

2. RLHFlow/Reinforce-Ada RLHFlow/Reinforce-Ada Public

   An adaptive sampling framework for Reinforce-style LLM post training.

   Python 92 16

3. frac-cot frac-cot Public

   An efficient sampling method for long-CoT LLM with fractured CoT.

   Python 16

4. RSD RSD Public

   [ICML 2025] Reward-guided Speculative Decoding (RSD) for efficiency and effectiveness.

   Python 56 5

5. ApiQ ApiQ Public

   [EMNLP 2024] Quantize LLM to extremely low-bit, and finetune the quantized LLMs

   Python 15 2

6. mefts mefts Public

   [NeurIPS 2023] Make Your Pre-trained Model Reversible: From Parameter to Memory Efficient Fine-Tuning

   Python 33 1