AI RESEARCH
Learning to Rank Caption Chains for Video-Text Alignment
arXiv CS.LG
•
ArXi:2603.25145v1 Announce Type: cross Direct preference optimization (DPO) is an effective technique to train language models to generate preferred over dispreferred responses. However, this binary "winner-takes-all" approach is suboptimal for vision-language models whose response quality is highly dependent on visual content. In particular, a response may still be faithful to the visual inputs even if it is less preferable than an alternative.