Metraipolis › archive › dc_c4a901c8dec4

Lost with a Map: Conversational State and Behavioral Reliability in Language Models — arXiv:2609.33883

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33883
License (as archived)
CC0-1.0
Archived
2026-10-07T21:53:11.691Z
SHA-256 of the archived text
6c40435cd98f146cfbca525da4fbd8eef1379cf1bd52834ea32cc3e6cebcfb0c
Stable URL of this page
https://metraipolis.com/documento/dc_c4a901c8dec4

How to cite

Lost with a Map: Conversational State and Behavioral Reliability in Language Models — arXiv:2609.33883 — arXiv (en), https://arxiv.org/abs/2609.33883 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c4a901c8dec4 (Metraipolis archive dc_c4a901c8dec4, sha256 6c40435cd98f146cfbca525da4fbd8eef1379cf1bd52834ea32cc3e6cebcfb0c)
BibTeX
@misc{dcc4a901c8dec4,
  title = {Lost with a Map: Conversational State and Behavioral Reliability in Language Models — arXiv:2609.33883},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33883},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c4a901c8dec4; sha256 6c40435cd98f146cfbca525da4fbd8eef1379cf1bd52834ea32cc3e6cebcfb0c},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33883. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Task-oriented dialogue requires maintaining and updating information across turns, yet language models expose no explicit belief-state object. We study how conversational state is represented, updated, and used inside eight instruction-tuned language models from four families on MultiWOZ and SGD. Structure and values separate: which domains, slots, and requests are active is linearly readable just before the model acts, whereas exact values are far more readable where the user stated them. After a user changes a value, both values remain accessible at their mentions, and causal interventions show that both continue to influence the model's action. In natural closed-loop interaction, query failures separate into cases of weak structural support, incorrect value resolution, and failure to deploy otherwise-supported constraints, with targeted interventions producing systematically different repair behavior across these cases. These findings motivate a state-action controller that starts from the base model action and selectively edits it using structural readouts, without requiring a complete predicted belief state as an intermediate representation. On held-out MultiWOZ interaction across five models, it raises the base model exact-query accuracy from .318 to .621 and task success from .272 to .371 at negligible added cost. Overall, reliable interaction requires not only retaining conversational information, but resolving which available constraints currently apply and ensuring that they govern action.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.33883 — 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 2026-10-07T21:27:08.560Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).