如何正确犯错：一个用于构建和基准测试错误感知第一人称视频的框架
How to Correctly Make Mistakes: A Framework for Constructing and Benchmarking Mistake Aware Egocentric Procedural Videos
AI总结：
本文提出PIE-V框架，通过引入可控的人类合理偏差来构建和评估错误感知的第一人称视频，结合心理学启发的错误计划、纠正计划、LLM作家和评委，提升视频的合理性和一致性。
摘要：
可靠的视频过程监控需要暴露自然发生的人类错误及其后续恢复。在第一人称记录中，错误常被手部部分遮挡并通过微妙的对象状态变化揭示，而现有过程数据集提供的错误和纠正轨迹有限且不一致。我们提出了PIE-V（心理启发的视频错误注入），一个通过在干净的步骤过程上添加受控、人类合理的偏差来构建和评估错误感知的第一人称视频的框架。PIE-V结合了一个基于过程阶段和语义步骤负载的心理学启发错误计划，一个建模恢复行为的纠正计划，一个执行级联一致重写的LLM作家，以及一个验证过程连贯性和修复失败的LLM评委。对于视频片段编辑，PIE-V通过文本引导的视频生成合成替换片段，并将其缝合到事件中以保持视觉合理性。应用于17个任务和50个Ego-Exo4D场景，PIE-V注入了102个错误并生成了27个恢复纠正。在基准测试中，我们引入了一个统一的分类法和一个包含九个指标的人类评估标准，涵盖步骤级和过程级的质量，包括合理性、过程逻辑与标注者信心、状态变化连贯性以及文本与视频之间的联系。使用此协议，我们审核了几个现有资源，并在相同标准下将PIE-V与自由形式LLM生成基线进行比较。共同，该框架和评估标准支持在第一人称过程错误检测和纠正后的验证。
英文摘要：
Reliable procedural monitoring in video requires exposure to naturally occurring human errors and the recoveries that follow. In egocentric recordings, mistakes are often partially occluded by hands and revealed through subtle object state changes, while existing procedural datasets provide limited and inconsistent mistake and correction traces. We present PIE-V (Psychologically Inspired Error injection for Videos), a framework for constructing and benchmarking mistake-aware egocentric procedural videos by augmenting clean keystep procedures with controlled, human-plausible deviations. PIE-V combines a psychology-informed error planner conditioned on procedure phase and semantic step load, a correction planner that models recovery behavior, an LLM writer that performs cascade-consistent rewrites, and an LLM judge that validates procedural coherence and repairs failures. For video segment edits, PIE-V synthesizes replacement clips with text-guided video generation and stitches them into the episode to preserve visual plausibility. Applied to 17 tasks and 50 Ego-Exo4D scenarios, PIE-V injects 102 mistakes and generates 27 recovery corrections. For benchmarking, we introduce a unified taxonomy and a human rubric with nine metrics that cover step-level and procedure-level quality, including plausibility, procedure logic with annotator confidence, state change coherence, and grounding between text and video. Using this protocol, we audit several existing resources and compare PIE-V against a freeform LLM generation baseline under the same criteria. Together, the framework and rubric support post-completion verification for egocentric procedural mistake detection and correction.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
