EgoAffect-R1: Elimination-Guided Reinforcement Learning for Egocentric Embodied Emotional Reasoning
Open Access
- 9 November 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 17517-17524
Abstract
Egocentric social-affective understanding is central to embodied affective intelligence, yet remains challenging for multimodal large language models because correct interpretations often depend on subtle egocentric evidence and fine-grained distinctions among plausible candidates. We propose EgoAffect-R1, a process-aware and elimination-guided reinforcement learning framework for egocentric embodied emotional reasoning. We construct EgoAffect-CoT-18k for cold-start CoT supervision and EgoAffect-RL-5k for verifiable reinforcement learning. Our EgoAffect reward scores format, question type, emotional tone, and elimination-based candidate discrimination, penalizing trajectories that discard the correct semantic option. To reduce option-order shortcuts, we introduce Option-Perturbed DAPO, which trains on reordered candidates and aligns rewards with their semantic identities. Experiments on EgoLink show that EgoAffect-R1 achieves 92.75% accuracy on the EgoLink validation split and 92.46% accuracy in the EgoLink 2026 Challenge. These results show that structured cold-start data, elimination-guided rewards, and option-perturbed RL improve egocentric affective reasoning. Code and data are available at https://github.com/ZhengQinLai/EgoAffect-R1.Keywords
Funding Information
- National Key R&D Program of China (2025YFC3811300)
- National Natural Science Foundation of China (62376070)
- National Natural Science Foundation of China (62076195)
This publication has 16 references indexed in Scilit:
- Incremental Micro-Expression Recognition: A BenchmarkIEEE Transactions on Affective Computing, 2026
- Agent-MER: A Cognitive Agent with Hierarchical Deliberation for Open-Vocabulary Multimodal Emotion Recognition2025
- Which of These Best Describes Multiple Choice Evaluation with LLMs? A) Forced B) Flawed C) Fixable D) All of the Above2025
- Addressing Blind Guessing: Calibration of Selection Bias in Multiple-Choice Question Answering by Video Language Models2025
- Multimodal Blockwise Transformer for Robust Sentiment Recognition2024
- Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questions2024
- POE: Process of Elimination for Multiple Choice Reasoning2023
- Ego4D: Around the World in 3,000 Hours of Egocentric Video2022
- Exploring the Contextual Factors Affecting Multimodal Emotion Recognition in VideosIEEE Transactions on Affective Computing, 2021
- Context-Aware Emotion Recognition Networks2019