English

RB-FT: Rationale-Bootstrapped Fine-Tuning for Video Classification

Computer Vision and Pattern Recognition 2025-11-21 v1

Abstract

Vision Language Models (VLMs) are becoming increasingly integral to multimedia understanding; however, they often struggle with domain-specific video classification tasks, particularly in cases with limited data. This stems from a critical \textit{rationale gap}, where sparse domain data is insufficient to bridge the semantic distance between complex spatio-temporal content and abstract classification labels. We propose a two-stage self-improvement paradigm to bridge this gap without new annotations. First, we prompt the VLMs to generate detailed textual rationales for each video, compelling them to articulate the domain-specific logic. The VLM is then fine-tuned on these self-generated rationales, utilizing this intermediate supervision to align its representations with the nuances of the target domain. Second, conventional supervised fine-tuning (SFT) is performed on the task labels, achieving markedly higher effectiveness as a result of the model's pre-acquired domain reasoning. Extensive experiments on diverse datasets demonstrate that our method significantly outperforms direct SFT, validating self-generated rationale as an effective, annotation-efficient paradigm for adapting VLMs to domain-specific video analysis.

Keywords

Cite

@article{arxiv.2511.15923,
  title  = {RB-FT: Rationale-Bootstrapped Fine-Tuning for Video Classification},
  author = {Meilong Xu and Di Fu and Jiaxing Zhang and Gong Yu and Jiayu Zheng and Xiaoling Hu and Dongdi Zhao and Feiyang Li and Chao Chen and Yong Cao},
  journal= {arXiv preprint arXiv:2511.15923},
  year   = {2025}
}

Comments

11 pages, 2 figures

R2 v1 2026-07-01T07:46:19.126Z