幻觉校正能否改善视频-语言对齐？
Can Hallucination Correction Improve Video-Language Alignment?
AI总结：
针对大型视觉-语言模型的幻觉问题，该研究提出自训练框架HACA，将幻觉校正作为训练目标，通过识别校正视频描述与内容的不一致来增强时空推理下的视频-语言对齐能力，在相关任务中取得稳定提升。
摘要：
大型视觉-语言模型常会生成未基于其视觉输入的幻觉内容。现有研究多聚焦于缓解幻觉，而我们转而探索将幻觉校正作为训练目标以提升视频-语言对齐效果。我们提出HACA（一种自训练框架），用于学习校正与视频内容不匹配的描述中的幻觉内容。通过识别并校正不一致之处，HACA增强了模型对齐视频与文本表征以进行时空推理的能力。实验结果显示，该方法在视频-字幕绑定和文本到视频检索任务中均取得了稳定提升，证明受幻觉校正启发的任务是改善视觉与语言对齐的有效策略。
英文摘要：
Large Vision-Language Models often generate hallucinated content that is not grounded in its visual inputs. While prior work focuses on mitigating hallucinations, we instead explore leveraging hallucination correction as a training objective to improve video-language alignment. We introduce HACA, a self-training framework learning to correct hallucinations in descriptions that do not align with the video content. By identifying and correcting inconsistencies, HACA enhances the model's ability to align video and textual representations for spatio-temporal reasoning. Our experimental results show consistent gains in video-caption binding and text-to-video retrieval tasks, demonstrating that hallucination correction-inspired tasks serve as an effective strategy for improving vision and language alignment.
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。
