PACE：因子引导的长视频证据获取框架

AI 摘要

香港科技大学团队提出PACE框架，用于长视频问答中的证据获取。PACE采用两阶段方法：先基于问题因子索引片段描述，再结合候选答案进行对比验证。在MMR-V基准上，PACE达到42.6%准确率，优于Deep Video Discovery等基线，并在多个长视频基准上表现一致。

正文节选

Finding the Right Evidence: Factor-Guided Coarse-to-Fine Reasoning for Long Videos Abstract While LVLMs rapidly improve, long-video question answering still remains challenging: relevant evidence is sparse, and question-relevant context often fails to provide cues that discriminate the correct answer from plausible alternatives. Diagnostic analysis on a manually annotated subset of MMR-V shows that prior agentic systems substantially improve cue retrieval over direct VLM inference yet fail to a

发布时间：2026-08-29 12:00

抓取时间：2026-08-28 18:29

来源机构：arXiv