Decoding Affective Nuances: Enhancing MLLMs via Hierarchical Emotion Reasoning and Contrastive Discriminative Pruning

The paper introduces DAN, a training‑free inference‑time framework that improves affective reasoning in multimodal large language models. It combines a Hierarchical Emotional Reasoning Chain (HERC) to better capture fine‑grained visual cues and a Contrastive Discriminative Visual Pruning (CDVP) module to isolate discriminative tokens for semantically similar emotions. Experiments show significant gains, notably a +10.47% improvement on the WebEmo25 benchmark with Qwen3‑VL‑8B‑Instruct.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computer Vision.

• llms
• efficiency
• multimodal
• benchmarks