3 papers
cs.AI2026
Measuring the Reasoning-Action Coupling in Driving Vision-Language-Action Models through Counterfactuals
Gabriel Manalu, Saad Wazir, Thao Do +4
Driving vision-language-action (VLA) models bring an appealing form of explainability to a safety-critical domain, since a system that can state the reasons for its maneuvers offer…
cs.CV2026
TB-AVA: Text as a Semantic Bridge for Audio-Visual Parameter Efficient Finetuning
Seongah Kim, Dinh Phu Tran, Hyeontaek Hwang +3
Audio-visual understanding requires effective alignment between heterogeneous modalities, yet cross-modal correspondence remains challenging when temporally aligned audio and visua…
cs.CL2026
Model-Dowser: Data-Free Importance Probing to Mitigate Catastrophic Forgetting in Multimodal Large Language Models
Hyeontaek Hwang, Nguyen Dinh Son, Daeyoung Kim
Fine-tuning Multimodal Large Language Models (MLLMs) on task-specific data is an effective way to improve performance on downstream applications. However, such adaptation often lea…