Metraipolis › archive › dc_4afc6850ce02

RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation — arXiv:2610.10120

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.10120
License (as archived)
CC0-1.0
Archived
2026-10-08T11:09:47.544Z
SHA-256 of the archived text
f3d954940940d62044a09b09c0dbacbf46f882baf368249e7871cc391178be15
Stable URL of this page
https://metraipolis.com/documento/dc_4afc6850ce02

How to cite

RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation — arXiv:2610.10120 — arXiv (en), https://arxiv.org/abs/2610.10120 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4afc6850ce02 (Metraipolis archive dc_4afc6850ce02, sha256 f3d954940940d62044a09b09c0dbacbf46f882baf368249e7871cc391178be15)
BibTeX
@misc{dc4afc6850ce02,
  title = {RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation — arXiv:2610.10120},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.10120},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4afc6850ce02; sha256 f3d954940940d62044a09b09c0dbacbf46f882baf368249e7871cc391178be15},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.10120. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in domains where task outcomes can be reliably evaluated, but long-horizon interaction remains challenging due to sparse terminal feedback and difficult credit assignment. Process rewards provide denser supervision, yet the capabilities most relevant for training can change as the policy evolves: a behavior that is easy to evaluate or frequently deficient need not be the bottleneck currently limiting task success. We introduce RewardWeaver, a self-evolving reward adaptation framework for language agents in long-horizon interaction. RewardWeaver maintains a validated capability space in which the semantics of admitted Rubrics remain fixed, and closes the loop between policy optimization, task evaluation, failure attribution, and reward adaptation. After each training stage, it performs outcome-grounded backward attribution on low-outcome trajectories, aggregates recurrent and policy-controlled capability bottlenecks, and dynamically selects the corresponding process rewards for the next stage. Recurrent failures not covered by the existing capability space trigger a separate, controlled expansion procedure. We evaluate REWARDWEAVER on SOTOPIA, Amazon?HistoryPrice, and a newly constructed Sales Benchmark. Across social interaction, bilateral bargaining, and domain-specific sales, REWARDWEAVER establishes new state-of-the-art (SOTA) results. Ablations further demonstrate the importance of dynamic reward allocation, failure-grounded attribution, and stable semantics for admitted capabilities.

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/2610.10120 — 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 2026-10-08T10:14:05.975Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).