AMPS: Adaptive Modality Preference Steering via Functional Entropy
arXiv:2602.12533v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) often exhibit significant modality preference, which is a tendency to favor one modality over another. …
Zihan Huang, Xintong Li, Rohan Surana, Tong Yu, Rui Wang, Julian McAuley, Jingbo Shang, Junda Wu
10 views