用偏好优化让视频字幕更符合人类关注点
AVC-DPO: Aligned Video Captioning via Direct Preference Optimization
- 通过设计针对性提示，聚焦时空信息增强人类偏好对齐
- 在VDC挑战赛中以最高分夺冠，显著提升字幕质量
- 适合需要精准描述视频内容的场景应用
尽管视频多模态大模型在视频字幕生成任务上已取得显著进展，但难以根据人类偏好调整字幕关注重点。为此，我们提出基于直接偏好优化的对齐视频字幕方法（AVC-DPO），一种后训练框架，通过偏好对齐增强视频多模态大模型的字幕生成能力。该方法设计了针对时间动态和空间信息的增强提示，这两者是人类观看视频时重点关注的因素，从而融入以人为本的偏好。AVC-DPO利用同一基础模型在不同提示条件下的生成响应，开展感知偏好的训练与字幕对齐。使用该框架，我们在CVPR'25 Workshop Track 1A：视频详细字幕挑战赛中表现优异，于视频详细字幕（VDC）基准测试中以最高分获得第一名，依据VDCSCORE评估指标。
原文摘要 · Abstract (English)
Although video multimodal large language models (video MLLMs) have achieved substantial progress in video captioning tasks, it remains challenging to adjust the focal emphasis of video captions according to human preferences. To address this limitation, we propose Aligned Video Captioning via Direct Preference Optimization (AVC-DPO), a post-training framework designed to enhance captioning capabilities in video MLLMs through preference alignment. Our approach designs enhanced prompts that specifically target temporal dynamics and spatial information-two key factors that humans care about when watching a video-thereby incorporating human-centric preferences. AVC-DPO leverages the same foundation model's caption generation responses under varied prompt conditions to conduct preference-aware training and caption alignment. Using this framework, we have achieved exceptional performance in the LOVE@CVPR'25 Workshop Track 1A: Video Detailed Captioning Challenge, achieving first place on the Video Detailed Captioning (VDC) benchmark according to the VDCSCORE evaluation metric.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。