CASE: Confusion-Aware Skill Evolution for Frozen Multimodal Emotion Models

Notably, CASE improves the average accuracy of Qwen3-VL-8B on the CA-MER benchmark from 46.80% to 59.13%, while achieving consistent gains across MLLMs. It comes from a paper posted to arXiv on 26 October 2026. Multimodal large language models (MLLMs) have substantially advanced Multimodal Emotion Recognition (MER) by enabling fine-grained emotion understanding through reasoning over multimodal evidence. However, they still exhibit systematic confusion between semantically adjacent emotions with highly overlapping multimodal cues. Existing approaches typically address these errors through global parameter adaptation, making confusion correction costly and inflexible. We instead observe that emotion confusion is highly localized and therefore better addressed through reusable pair-specific expertise. Building on this insight, we propose Confusion-Aware Skill Evolution (CASE), the first agent-skill framework for MER. CASE maintains a sparse, self-evolving library of pair-specific EmotionSkills that externalize affective expertise from model parameters, enabling the learned expertise to be continuously refined and readily transferred across diverse frozen MLLM backbones. During inference, CASE dynamically retrieves the appropriate EmotionSkill to reinforce boundary-aware reasoning while keeping the MLLM backbone frozen. Extensive experiments on CA-MER, DFEW, and EMER demonstrate the effectiveness and transferability of CASE across both general-purpose and emotion-specialized MLLMs.

Why it counts

Notably, CASE improves the average accuracy of Qwen3-VL-8B on the CA-MER benchmark from 46.80% to 59.13%, while achieving consistent gains across MLLMs. Building on this insight, we propose Confusion-Aware Skill Evolution (CASE), the first agent-skill framework for MER.

Sources

Primary source: primary source

What is not known

This brief does not claim independent replication. Claims that appear only on X and not in the primary source stay unknown.