通过候选感知因果推理增强教学视频中的时间答案定位
CACR:Reinforcing Temporal Answer Grounding in Instructional Video via Candidate-Aware Causal Reasoning
- University of Science and Technology of China（中国科学技术大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出候选感知因果推理框架，通过视觉-语言预训练候选选择和基于GRPO的时序逻辑推理，解决教学视频中复杂问题理解和长视频片段定位挑战，在六个基准上取得最优mIoU。
AI中文摘要：
教学视频中的时间答案定位任务旨在定位响应自然语言查询的精确视频片段，对于直接视频答案检索日益重要。由于需要理解语义复杂的问题并解决未修剪视频与短目标时刻之间的显著长度不匹配，该任务仍然具有挑战性。现有方法通常对无关内容敏感或视觉推理能力不足。为了解决这些局限性，我们提出了候选感知因果推理框架。我们的方法首先采用基于视觉-语言预训练的候选选择算法高效生成K个候选片段，然后应用由拒绝奖励机制增强并通过组相对策略优化优化的时序逻辑推理模块进行稳健推理。在六个基准上的大量实验表明，我们的方法在平均交并比方面达到了最先进的性能，为长视频中基于推理的检索提供了新视角。
英文摘要：
The task of temporal answer grounding in instructional video (TAGV), which aims to locate precise video segments that respond to natural language queries, is increasingly important for direct video answer retrieval. This task remains challenging due to the need to comprehend semantically complex questions and to address the significant length mismatch between untrimmed videos and short target moments. Existing methods often suffer from sensitivity to irrelevant content or insufficient visual reasoning capabilities. To tackle these limitations, we propose a Candidate-Aware Causal Reasoning (CACR) framework. Our approach first employs a Visual-Language Pre-training based Candidate Selection (VBCS) algorithm to efficiently generate K candidate segments, then applies a temporal logic reasoning module enhanced by a rejection reward mechanism and optimized via Group Relative Policy Optimization (GRPO) for robust inference. Extensive experiments on six benchmarks demonstrate that our method achieves state-of-the-art performance in terms of mean Intersection-over-Union (mIoU), providing a new perspective for reasoning-based retrieval in long videos.
