Popular repositories Loading

1. hal-harness hal-harness Public

   Python 278 55

2. RLMT RLMT Public

   [R]einforcement [L]earning from [M]odel-rewarded [T]hinking - code for the paper "Language Models That Think, Chat Better"

   Python 129 6

3. MeCo MeCo Public

   Code for ICML 25 paper "Metadata Conditioning Accelerates Language Model Pre-training (MeCo)"

   Python 51 2

4. PruLong PruLong Public

   Code for the preprint "Cache Me If You Can: How Many KVs Do You Need for Effective Long-Context LMs?"

   Python 48 4

5. what-makes-good-rm what-makes-good-rm Public

   [NeurIPS 2025] What Makes a Reward Model a Good Teacher? An Optimization Perspective

   Python 43 4

6. retaining-by-doing retaining-by-doing Public

   Python 40 3