18 papers
TREK: Distill to Explore, Reinforce to Refine
Yuanda Xu, Zhengze Zhou, Kayhan Behdin +10
Group Relative Policy Optimization (GRPO) is effective when the current policy already samples useful reasoning trajectories, but it stalls on hard prompts whose correct solution m…
TRIAGE: Role-Typed Credit Assignment for Agentic Reinforcement Learning
Yuanda Xu, Zhengze Zhou, Hejian Sang +6
Agentic reinforcement learning requires assigning credit to environment-facing actions such as searches, clicks, edits, navigation commands, and object interactions. Standard GRPO…
CRISP: Compressed Reasoning via Iterative Self-Policy Distillation
Hejian Sang, Yuanda Xu, Zhengze Zhou +3
Reasoning models often generate far more tokens than a task requires, which raises inference cost and can compound errors. We introduce CRISP (Compressed Reasoning via Iterative Se…
Distilling the Essence: Efficient Reasoning Distillation via Sequence Truncation
Wei-Rui Chen, Vignesh Kothapalli, Ata Fatahibaarzi +5
Distilling the capabilities from a large reasoning model (LRM) to a smaller student model often involves training on substantial amounts of reasoning data. However, knowledge disti…
TIP: Token Importance in On-Policy Distillation
Yuanda Xu, Hejian Sang, Zhengze Zhou +3
On-policy knowledge distillation (OPD) trains a student on its own rollouts under token-level supervision from a teacher. Not all token positions matter equally, but existing views…
Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training
Yuanda Xu, Hejian Sang, Zhengze Zhou +3
In settings where labeled verifiable training data is the binding constraint, each checked example should be allocated to the model and reward density where it is most informative.…