This project investigates whether curriculum-based sampling can improve RLOO training by concentrating updates on prompts that are expected to be most informative and suggests a fundamental tradeoff between concentrating updates on potentially informative prompts and maintaining broad coverage of the training distribution.
This work proposes RLPF, reinforcement learning from performance feedback, which turns execution outcomes into a staged reward, and suggests that code agents can be trained not only to pass tests, but also to optimize the programs they write.
Huihao Jing, Hao-Zhe Cui, Wenbin Hu et al.· arXiv.org· 0 citations
This work proposes two complementary strategies to improve the performance of value function RL: Privileged Value Functions (PVF) which provide an elegant mechanism to inject additional task-relevant token-level signal without biasing the policy objective; and TETHER, a baseline that adaptively interpolates between gro...
S. Venkatraman, Matthieu Dinot, Laurence Aitchison· 1 citation· ⚡1
Off-Context GRPO (OC-GRPO), a minimally modified variant of GRPO that uses guided rollouts but applies an importance-corrected objective to steer the update back toward the original unguided objective, avoiding the mismatch that destabilizes uncorrected guided training.
Priyank Agrawal, Ankur Samanta, S. Ghasemlou et al.· arXiv.org· 1 citation
Direct On-Policy Distillation (Direct-OPD) is proposed, which transfers the teacher's RL-induced policy shift instead of running sparse-reward RL on the target model and consistently leverages weaker teachers to improve stronger target models.
Shiyuan Feng, Huan Gao, Haohan Chi et al.· arXiv.org· 8 citations· ⚡1
TrajVal, a lightweight probe-based estimator that approximates per-task learnability from a short probe run and two endpoint evaluations, is proposed and it is found that learnability is reproducible across independently sampled training contexts and predictive of downstream utility.
Ting Zhou, Zhenqing Ling, Dao-Yuan Chen et al.· 1 citation
Desc descriptive evidence is provided that long-horizon multi-tool post-training can change ways of working that transfer beyond its training domain, and both software-engineering benchmarks improve despite the training collection containing no software-engineering tasks.
Sushant Mehta, Logan Ritchie, Liudas Panavas et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
