Pinned Loading

1. LUFFY LUFFY Public

   Official Repository of "Learning to Reason under Off-Policy Guidance"

   Python 459 69

2. TPO TPO Public

   Test-time preferenece optimization (ICML 2025).

   Jupyter Notebook 185 12

3. SU-01 SU-01 Public

   SU-01: Achieving Gold-Medal-Level Olympiad Reasoning via Simple and Unified Scaling

   Python 94 5

4. Pi-Bench Pi-Bench Public

   Benchmark for proactive personal assistant agents in long-horizon workflows.

   Python 52 1

5. TRM TRM Public

   Code repository for the ICML 2026 Oral paper "Characterizing, Evaluating, and Optimizing Complex Reasoning".

   Python 17