研究人员推出了一种新颖的MultAttnAttrib方法，用于在多模态问答系统中生成归因，而无需额外训练。该方法利用模型的预填充通道、特定的注意力头和校准的阈值来精确定位文档中的证据。为了评估其有效性，创建了一个名为MultAttrEval的新基准数据集，其中包含基于多模态来源的答案的细粒度归因。MultAttnAttrib在现有归因方法（包括基于提示的方法）方面表现出色，甚至能与GPT 5.4等先进模型相媲美，同时显著降低了推理延迟。 AI
影响 通过提高答案归因的准确性和效率，增强了基于QA系统的信任和安全性。
排序理由 该集群描述了一篇介绍用于问答中多模态归因的新颖方法和数据集的研究论文。[lever_c_demoted from research: ic=1 ai=1.0]
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2607.01420v1 Announce Type: cross Abstract: As grounded QA systems are increasingly deployed in AI assistants, accurately attributing generated answers to evidence is critical for user trust and model safety. While unimodal attributions have been explored in depth, the mult…
As grounded QA systems are increasingly deployed in AI assistants, accurately attributing generated answers to evidence is critical for user trust and model safety. While unimodal attributions have been explored in depth, the multimodal setting remains relatively under-researched…