arXiv：cs.CL· Shuoyang Sun, Kerui Gu, Hao Fang, Shaoli Huang, Bin Chen·· 3 小时前
ReTurn：多模态多轮对话中历史信息的选择性使用评测
When History Helps and Hurts: Selective History Use across Multimodal Turns
AI 导读
研究团队推出 ReTurn 基准，包含 7000 个覆盖视觉与音频证据的基础任务，用于评测多模态模型在多轮对话中对历史信息的选择性使用。在 13 个全模态、视觉语言和音频语言模型上，中位开放式准确率从直接输入的 93.7% 降至对话中的 72.3%。行为探测显示高问题回忆率可能与较弱任务应用并存，竞争媒体会将答案从历史目标引开，监督适配仅带来部分提升。
正文
Abstract: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on selective use of conversational history: an earlier question may remain relevant while its previous answer is outdated, whereas a current request may depend on historical evidence despite conflicting new observations.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.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.11948 [cs.CL] |
| (or arXiv:2610.11948v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.11948 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Shuoyang Sun [view email]
[v1]
Thu, 8 Oct 2026 13:36:12 UTC (1,534 KB)
来源：arXiv：cs.CL · arxiv.org