Metraipolis › archive › dc_f38ef4469ddd

IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents — arXiv:2608.06735

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.06735
License (as archived)
CC0-1.0
Archived
2026-10-19T09:47:07.242Z
SHA-256 of the archived text
aa8b2d30c8f928c2dd53e57f4ad45cbb68faedb3d7e2ce4b871bede00cc4be53
Stable URL of this page
https://metraipolis.com/documento/dc_f38ef4469ddd

How to cite

IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents — arXiv:2608.06735 — arXiv (en), https://arxiv.org/abs/2608.06735 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f38ef4469ddd (Metraipolis archive dc_f38ef4469ddd, sha256 aa8b2d30c8f928c2dd53e57f4ad45cbb68faedb3d7e2ce4b871bede00cc4be53)
BibTeX
@misc{dcf38ef4469ddd,
  title = {IB-RL: Isolated Bilateral Reinforcement Learning for Strategic Dialogue Agents — arXiv:2608.06735},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.06735},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f38ef4469ddd; sha256 aa8b2d30c8f928c2dd53e57f4ad45cbb68faedb3d7e2ce4b871bede00cc4be53},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.06735. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reinforcement learning (RL) has achieved strong results in improving large language models (LLMs) on tasks with stationary, verifiable rewards, such as mathematical reasoning and code execution. In these settings, the environment follows fixed rules and does not adapt strategically to the agent. Strategic dialogue differs in this respect: the environment is another agent that adapts to the policy, and success depends on the interaction between the two sides. Despite this interactive nature, current RL approaches typically train a target agent against a fixed counterpart or simulator. We find that this training paradigm encourages the policy to exploit counterpart-specific regularities rather than learn strategies that generalize across counterparts. We call this problem the static-counterpart mismatch, which we quantify directly in our experiments. To address it, we propose Isolated Bilateral Reinforcement Learning (IB-RL), in which the two roles coevolve through joint rollouts while each role optimizes its own reward through fully independent advantages, action masks, and update paths. We evaluate frozen policies against fully independent held-out counterparts in both domains. On Vehicle TeleSales, IB-RL achieves 89.6% Success@1, compared to 84.6% for the best unilateral RL baseline. On Deal-or-NoDeal, it reaches 98.4% agreement against DeepSeek V4 Pro, compared to 86.4% for the best unilateral baseline. These results indicate that jointly training both roles with strict peragent isolation produces policies that generalize more effectively to unseen counterparts.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-10 in cs.CL).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.06735 — 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 2026-10-19T09:46:37.261Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-08-10?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).