Metraipolis › archive › dc_0a410081a86a

EliteReplay: Preserving Diverse Solutions in RL Post-Training via Quality-Diversity Search — arXiv:2610.22290

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.22290
License (as archived)
CC0-1.0
Archived
2026-10-16T09:05:50.607Z
SHA-256 of the archived text
344a9845ea6e325ec4827c6709d16e0bbb4b5a126cf57b6eaeef4ae1bcc756d9
Stable URL of this page
https://metraipolis.com/documento/dc_0a410081a86a

How to cite

EliteReplay: Preserving Diverse Solutions in RL Post-Training via Quality-Diversity Search — arXiv:2610.22290 — arXiv (en), https://arxiv.org/abs/2610.22290 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_0a410081a86a (Metraipolis archive dc_0a410081a86a, sha256 344a9845ea6e325ec4827c6709d16e0bbb4b5a126cf57b6eaeef4ae1bcc756d9)
BibTeX
@misc{dc0a410081a86a,
  title = {EliteReplay: Preserving Diverse Solutions in RL Post-Training via Quality-Diversity Search — arXiv:2610.22290},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.22290},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_0a410081a86a; sha256 344a9845ea6e325ec4827c6709d16e0bbb4b5a126cf57b6eaeef4ae1bcc756d9},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22290. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning has become a common post-training technique to improve the accuracy of LLMs on verifiable tasks. However, such training can narrow the set of solutions the model can produce. Several approaches have attempted to address this behavior-sharpening effect, but they often optimize metrics that do not reflect the diversity of correct solutions. Ideally, we want a model that can produce multiple good solutions, not only a single reliable one. We propose EliteReplay, a post-training recipe that pairs GRPO with an "elite" archive built by quality-diversity search. Before training, we use the frozen base model to collect solutions that span distinct approaches; during training, GRPO is combined with a replay loss on this archive. The result is a post-trained model that improves accuracy while preserving more of the base model's solution diversity. On multiple text-to-SQL tasks, EliteReplay achieves a higher pass@128 than all baselines, producing up to 77% more distinct correct strategies than the strongest baselines. It preserves most of the base model's correct strategies, outperforming existing methods in strategy retention. This diversity translates into practical benefits on ambiguous questions, where EliteReplay covers both interpretations nearly twice as often as the strongest baseline on AmbiQT.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.22290 — 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 2026-10-16T08:44:17.835Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).