Metraipolis › archive › dc_ceb8cfe1ccfa

Learning to Reason and Use Tools through Unsupervised Fine-Tuning in Task-Oriented Dialog Systems — arXiv:2608.30426

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.30426
License (as archived)
CC0-1.0
Archived
2026-10-17T08:07:33.273Z
SHA-256 of the archived text
6adafa63523a388e5bb4d35b0a657b3ad9b63e6f284fc112726e9d3babd76495
Stable URL of this page
https://metraipolis.com/documento/dc_ceb8cfe1ccfa

How to cite

Learning to Reason and Use Tools through Unsupervised Fine-Tuning in Task-Oriented Dialog Systems — arXiv:2608.30426 — arXiv (en), https://arxiv.org/abs/2608.30426 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ceb8cfe1ccfa (Metraipolis archive dc_ceb8cfe1ccfa, sha256 6adafa63523a388e5bb4d35b0a657b3ad9b63e6f284fc112726e9d3babd76495)
BibTeX
@misc{dcceb8cfe1ccfa,
  title = {Learning to Reason and Use Tools through Unsupervised Fine-Tuning in Task-Oriented Dialog Systems — arXiv:2608.30426},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.30426},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ceb8cfe1ccfa; sha256 6adafa63523a388e5bb4d35b0a657b3ad9b63e6f284fc112726e9d3babd76495},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.30426. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Current dialogue systems struggle with dynamic information retrieval, often leading to hallucinations and lower response accuracy. We address this by adapting the ReAct framework for Task-Oriented Dialogue, enabling Large Language Models (LLMs) to access external knowledge and produce factual responses. Mainly, we propose an unsupervised fine-tuning pipeline that harvests reasoning trajectories via in-context learning inference. High-quality samples are filtered using an LLM-based judge to construct a robust training set. This is enhanced by a unsupervised self-improvement loop, where improved checkpoints generate increasingly better trajectories for subsequent fine-tuning iterations. Experiments on the SIMMC dataset demonstrate that ReAct-based systems outperform baselines due to superior reasoning and tool use. Notably, our fine-tuned 8B model surpasses a 70B in-context system. Finally, we present an error analysis, impact of scene complexity, and cross-domain generalization.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2608.30426 — 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 2026-10-17T07:47:52.815Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).