cs.CVDate pending

Video2Reaction: Training Foundation Video Models to Predict Audience Reaction

Authors: Sidong ZhangTrang NguyenShiv ShankarGauri JagatapDeepak ChandranAndrea FanelliMadalina Fiterau

Abstract

We introduce Video2Reaction, a multimodal dataset that maps short movie segments to the induced emotional reactions of viewers in the wild, as expressed through social media comments. Video2Reaction captures the natural diversity of emotional responses by aggregating reactions from online comments at scale, modeling labels as distributions over categorical emotions to better reflect the subjective and ambiguous nature of emotional perception. We benchmark two vision-language models (VLMs) finetuned with LoRA, showing that VLMs learn effectively from Video2Reaction and outperform specialized baselines on dominant reaction prediction. We further demonstrate that VLMs pre-finetuned on Video2Reaction transfer effectively to VCE, another induced emotion dataset with a different taxonomy and video domain. Notably, LLaVA-NeXT-Video-7B pre-finetuned on Video2Reaction and adapted on only 1% of VCE training data achieves a top-3 accuracy of 0.682, on par with the best reported VCE performance trained on the full dataset. The dataset is available at https://huggingface.co/datasets/infofusionlab/Video2Reaction

Explore similar work

CardsList
  1. Benchmarking Dynamic Affective Reasoning: A Viewer-Centric Video Emotion Dataset

    Jul 11, 2026Zhiyan Zhang, Peipei Song, Jinpeng Hu +3