arxiv
PublishedSeptember 30, 2026 at 4:00 AM
Learning Beyond What You Sample: Off-Policy-Aware Cross-Model Trajectory Exchange for RLVR
Publisher summary· verbatim
arXiv:2609.37868v1 Announce Type: cross Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) methods such as GRPO rely on successful self-generated trajectories, but finite rollout budgets can produce all-fail groups with no reward-based policy-gradient signal. While additional rollouts i
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivPredictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance2harxivPixel-Level Transformers in Remote Sensing: A Canopy Height Case Study2harxivExplore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents2harxivBoosting Adversarial Robustness and Generalization with Dictionary Structure2hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗