Video-Text Dataset Construction from Multi-AI Feedback: Promoting Weak-to-Strong Preference Learning for Video Large Language Models

High-quality video-text preference data is crucial for Multimodal Large Language Models (MLLMs) alignment. However, existing preference data is very scarce. Obtaining VQA preference data for preference training is costly, and manually annotating responses is highly unreliable, which could result in low-quality pairs. Meanwhile, AI-generated responses controlled by temperature adjustment lack diversity. To address these issues, we propose a high-quality VQA preference dataset, called textitMultiple Multimodal Artificial Intelligence Preference Datasets in VQA (MMAIP-V), which is constructed by sampling from the response distribution set and using an external scoring function for response evaluation. Furthermore, to fully leverage the preference knowledge in MMAIP-V and ensure sufficient optimization, we propose textitIterative Weak-to-Strong Reinforcement Learning from AI Feedback for video MLLMs (Iter-W2S-RLAIF), a framework that gradually enhances MLLMs' alignment capabilities by iteratively updating the reference model and performing parameter extrapolation. Finally, we propose an unbiased and information-complete evaluation scheme in VQA evaluation. Experiments demonstrate that MMAIP-V is beneficial for MLLMs in preference learning and Iter-W2S-RLAIF fully exploits the alignment information in MMAIP-V. We believe that the proposed automatic VQA preference data generation pipeline based on AI feedback can greatly promote future work in the MLLMs alignment. Code and dataset are available hrefhttps://anonymous.4open.science/r/MMAIP-VIter-W2S-RLAIF-702FMMAIP-VIter-W2S-RLAIF-702F.