Metraipolis › archive › dc_c007d09eda0a
When History Helps and Hurts: Selective History Use across Multimodal Turns — arXiv:2610.11948
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.11948
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T06:13:28.076Z
- SHA-256 of the archived text
be2f3defd8f226ae836b4e231f69d099e25a6264df4861ac0cef222ae7afacc8- Stable URL of this page
https://metraipolis.com/documento/dc_c007d09eda0a
How to cite
When History Helps and Hurts: Selective History Use across Multimodal Turns — arXiv:2610.11948 — arXiv (en), https://arxiv.org/abs/2610.11948 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c007d09eda0a (Metraipolis archive dc_c007d09eda0a, sha256 be2f3defd8f226ae836b4e231f69d099e25a6264df4861ac0cef222ae7afacc8)
BibTeX
@misc{dcc007d09eda0a,
title = {When History Helps and Hurts: Selective History Use across Multimodal Turns — arXiv:2610.11948},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.11948},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c007d09eda0a; sha256 be2f3defd8f226ae836b4e231f69d099e25a6264df4861ac0cef222ae7afacc8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11948. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reliable multimodal interaction depends on selective use of conversational history: an earlier question may remain relevant while its previous answer is outdated, whereas a current request may depend on historical evidence despite conflicting new observations. Existing multi-turn evaluations rarely separate these history-use demands from underlying question difficulty. To address this gap, we introduce ReTurn, a benchmark of 7,000 base tasks spanning visual and audio evidence for evaluating selective history use. For task-carrying history, Reconfirm/Reground require applying a historical question to current media while varying historical agreement; for evidence-carrying history, Retrieve/Rebind require answering a current question using historical media while varying current-media competition. Each pair preserves the target question, media, and answer. Tasks support open-ended and multiple-choice evaluation, with matched single-turn counterparts serving as answerability references. Across 13 omni-modal, vision-language, and audio-language models, median model-level open-ended accuracy falls from 93.7% with direct input to 72.3% in conversation. Behavioral probes show that high question recall can coexist with weaker task application, while competing media can redirect answers away from historical targets. Supervised adaptation yields only partial gains. ReTurn provides a controlled framework for assessing whether multimodal models select and use the historical information required by each request. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.11948 — 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 2026-10-09T05:47:15.308Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).