← Back to all news
arXiv AI By Chengsheng Zhang, Chenghao Sun, Zhining Xie, Xinmei Tian

Interpreting and Enhancing Emotional Circuits in Large Vision-Language Models via Cross-Modal Information Flow

Read the original on arXiv AI →

The paper introduces a steering‑vector‑based causal attribution framework to study how large vision‑language models (LVLMs) translate visual input into emotional narratives. By creating a specialized dataset, the authors uncover a functional decoupling in the LVLM’s three‑stage Adapt‑Aggregate‑Execute mechanism: visual emotional cues are first aggregated in middle layers via sentiment‑specific attention heads, then translated into narrative generation in deeper layers through emotion‑general pathways. Using these insights, they regulate emotional information routing to strengthen attention flow and amplify semantic activation, achieving significant performance gains on the MER‑UniBench and reducing emotional hallucinations through inference‑time intervention.

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 AI.

Related stories

arXiv Computation and Language
Sep 7

From Vision to Language: Investigating Causal Information Flow in Multimodal Decision-Making

The study examines how Vision‑Language Models (VLMs) integrate visual evidence into language‑based decisions by applying layer‑wise causal interventions on video‑text attention pathways in a video‑based generative multiple‑choice setting. Findings reveal that visual information is primarily incorporated while processing candidate answer options, with nouns serving as key semantic anchors and verbs becoming important during temporal reasoning. The research also uncovers a distinct pattern in temporal reasoning, indicating that VLMs struggle to reconstruct sequential information across video frames, possibly due to linguistic biases in temporal expressions.

By Davide Testa, Hugh Mee Wong, Alessandro Lenci, Bernardo Magnini, Albert Gatt
llmsmultimodal
More like this →
arXiv Computer Vision
4d ago

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.

By Cheng Ye, Weidong Chen, Zhaobo Qi, Beier Zhu, Zhendong Mao
llmsefficiencymultimodalbenchmarks
More like this →