Metraipolis › archive › dc_ab55b4e176a4

Dataset Mismatch Matters in Group Relative Policy Optimization for Reinforcement Learning from Verifiable Rewards — arXiv:2611.00586

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.00586
License (as archived)
CC0-1.0
Archived
2026-11-03T11:19:28.035Z
SHA-256 of the archived text
2f6fb5a2d96f057b44b779999b3abf9a38bb3984a18e0a79db6330a94e4106c0
Stable URL of this page
https://metraipolis.com/documento/dc_ab55b4e176a4

How to cite

Dataset Mismatch Matters in Group Relative Policy Optimization for Reinforcement Learning from Verifiable Rewards — arXiv:2611.00586 — arXiv (en), https://arxiv.org/abs/2611.00586 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ab55b4e176a4 (Metraipolis archive dc_ab55b4e176a4, sha256 2f6fb5a2d96f057b44b779999b3abf9a38bb3984a18e0a79db6330a94e4106c0)
BibTeX
@misc{dcab55b4e176a4,
  title = {Dataset Mismatch Matters in Group Relative Policy Optimization for Reinforcement Learning from Verifiable Rewards — arXiv:2611.00586},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00586},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ab55b4e176a4; sha256 2f6fb5a2d96f057b44b779999b3abf9a38bb3984a18e0a79db6330a94e4106c0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00586. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recently, reinforcement learning from verifiable rewards (RLVR) is practically very important, where the leading approach is group relative policy optimization (GRPO) that typically faces a trade-off between rollout efficiency and downstream performance. To improve efficiency while preserving performance, existing selective rollout methods mainly exploit only source-side training dynamics to optimize the discrete distribution over source prompts. However, in practice, the source training dataset is rarely identically distributed to the target test dataset. When such a mismatch exists, due to legal or privacy constraints, it is usually not acceptable to share target prompts themselves with the RLVR trainer. Nevertheless, it is still possible for the target data owner to provide limited aggregated feedback from a small held-out target set without revealing target prompts or rollout traces. In this paper, we propose learning-to-reweight rollout (L2RR) that explicitly incorporates target-side performance feedback into GRPO. Specifically, L2RR learns a discrete sampling distribution over source prompts by optimizing a bi-level objective that maximizes accuracy on a held-out target dataset, while simultaneously penalizing rollouts that provide little incremental learning information. Empirical studies on six math reasoning benchmarks and three model scales show that L2RR improves target-domain performance and increases rollout efficiency.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2611.00586 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-11-03T11:02:37.454Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).