# Video2Reaction: Mapping Video to Audience Reaction Distribution in the Wild

arXiv: https://arxiv.org/abs/2607.06875 · HTML: https://mlsanity.com/paper/2607.06875/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 38 references that are in the map.
- First posted on arXiv 8 Jul 2026.

## Builds on (most cited first, 22 of 22)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [Large-scale Contrastive Language-Audio Pretraining with Feature Fusion and Keyword-to-Caption Augmentation](https://mlsanity.com/paper/2211.06687/) (2022)
- [MELD: A Multimodal Multi-Party Dataset for Emotion Recognition in Conversations](https://mlsanity.com/paper/1810.02508/) (2018)
- [GoEmotions: A Dataset of Fine-Grained Emotions](https://mlsanity.com/paper/2005.00547/) (2020)
- [Visual Transformers: Token-based Image Representation and Processing for Computer Vision](https://mlsanity.com/paper/2006.03677/) (2020)
- [Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis](https://mlsanity.com/paper/2109.00412/) (2021)
- [Label Distribution Learning](https://mlsanity.com/paper/1408.6027/) (2014)
- [Aligning Language Models to User Opinions](https://mlsanity.com/paper/2305.14929/) (2023)
- [Condensed Movies: Story Based Retrieval with Contextual Embeddings](https://mlsanity.com/paper/2005.04208/) (2020)
- [Benchmarking Distributional Alignment of Large Language Models](https://mlsanity.com/paper/2411.05403/) (2024)
- [CubeMLP: An MLP-based Model for Multimodal Sentiment Analysis and Depression Estimation](https://mlsanity.com/paper/2207.14087/) (2022)
- [CommunityLM: Probing Partisan Worldviews from Language Models](https://mlsanity.com/paper/2209.07065/) (2022)
- [Exploiting Multi-Label Correlation in Label Distribution Learning](https://mlsanity.com/paper/2308.01742/) (2023)
- [Multi-News+: Cost-efficient Dataset Cleansing via LLM-based Data Annotation](https://mlsanity.com/paper/2404.09682/) (2024)
- [To Ensemble or Not: Assessing Majority Voting Strategies for Phishing Detection with Large Language Models](https://mlsanity.com/paper/2412.00166/) (2024)
- [How Would The Viewer Feel? Estimating Wellbeing From Video Scenarios](https://mlsanity.com/paper/2210.10039/) (2022)
- [Infer Induced Sentiment of Comment Response to Video: A New Task, Dataset and Baseline](https://mlsanity.com/paper/2407.06115/) (2024)

## Abstract (arXiv)

Understanding and forecasting audience reactions to video content are crucial for improving content creation, recommendation systems, and media analysis. To enable audience reaction prediction and other content engagement applications, we introduce Video2Reaction, a multimodal dataset that maps short movie segments to a distribution of textitinduced emotions of viewers in the wild, as expressed through social media. Video2Reaction spans more than 10,000 videos and serves as a reliable benchmark as well as a training resource for audience reaction prediction. To enable cost-effective continuous annotations as reactions may change over time, we develop a two-stage multi-agent pipeline using only open-source LLMs, achieving 86% correctness under blind human verification despite the inherently noisy and subjective nature of the task. We establish the first benchmark for video-to-reaction-distribution prediction in the wild and show that pretrained foundation video models fail in zero-shot settings, while finetuning transforms them into state-of-the-art predictors capable of modeling both full reaction distributions and dominant responses from video alone. However, the task remains challenging: even the strongest methods achieve only 77% Top-3 F1 in dominant reaction prediction (LLaVA-Next), highlighting a substantial gap in modeling collective audience reaction. modificationDataset and code are available at our project page: https://information-fusion-lab-umass.github.io/video2reaction-bench.github.io