作者位次 #2/4ACL 2026Findings· Findings of ACL
SCOPE : Preserving Modality-Specific Cues to Mitigate Modality Laziness in Multimodal Learning
Jingfan Yang · Rui Zhang · liang hong · Ke Yuan
摘要
Multimodal learning aims to learn unified multimodal representations from heterogeneous modalities and supports many natural language processing tasks. However, multimodal models often exhibit modality laziness: over-relying on a dominant modality and under-exploiting complementary signals. Exist...