Metraipolis › archive › dc_f3350a6db3cc

UnifiedPlayers: Enhance Tool-Integrated Reasoning in Agentic Reinforcement Learning — arXiv:2609.20089

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.20089
License (as archived)
CC0-1.0
Archived
2026-10-10T03:21:26.760Z
SHA-256 of the archived text
587f1f76b737c1e65938c60a3128a890573e308eeda87a28fee3147d05ae7249
Stable URL of this page
https://metraipolis.com/documento/dc_f3350a6db3cc

How to cite

UnifiedPlayers: Enhance Tool-Integrated Reasoning in Agentic Reinforcement Learning — arXiv:2609.20089 — arXiv (en), https://arxiv.org/abs/2609.20089 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f3350a6db3cc (Metraipolis archive dc_f3350a6db3cc, sha256 587f1f76b737c1e65938c60a3128a890573e308eeda87a28fee3147d05ae7249)
BibTeX
@misc{dcf3350a6db3cc,
  title = {UnifiedPlayers: Enhance Tool-Integrated Reasoning in Agentic Reinforcement Learning — arXiv:2609.20089},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.20089},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f3350a6db3cc; sha256 587f1f76b737c1e65938c60a3128a890573e308eeda87a28fee3147d05ae7249},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20089. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Self-evolving methods reduce the need for human-annotated trajectories by allowing tool-using agents to generate their own training data. Yet existing methods typically separate trajectory generation from evaluation, relying on static verifiers that cannot adapt to emerging failure modes or self-consistency signals that may reinforce errors shared across trajectories. Jointly adapting planning, execution, and evaluation offers a promising alternative, but introduces a fundamental coordination challenge: each component continuously changes the data or feedback used to train the others. We address this challenge with \textbf{UnifiedPlayers}, a cooperative framework comprising a Planning Player that generates tasks, an Execution Player that produces multi-turn trajectories with Python tool calls, and an Evaluation Player that constructs executable verifiers. We design role-specific rewards that coordinate the three players toward a shared learning objective under GRPO. Across two model backbones and twelve reasoning benchmarks, UnifiedPlayers outperforms the strongest prior baseline by at least 3.5\% on mathematical reasoning and 3.9\% on general reasoning tasks. Moreover, the learned verifier achieves 84.2\% adversarial detection accuracy, while its reward signal exhibits 2.03$\times$ higher per-question variance than a self-consistency baseline, providing more discriminative verifications. These results highlight cooperation among specialized players as a promising path toward self-enhanced tool-integrated agents.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.20089 — 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 2026-10-10T03:04:13.705Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).