CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion Paper • 2610.29915 • Published 7 days ago • 18
OmniReasoning: Pushing the Limits of Audio-Visual Joint Reasoning Paper • 2609.39490 • Published 28 days ago • 30
M^3Eval: Multi-Modal Memory Evaluation through Cognitively-Grounded Video Tasks Paper • 2606.05008 • Published Jun 3 • 26
AIM: Adaptive Inference of Multi-Modal LLMs via Token Merging and Pruning Paper • 2412.03248 • Published Dec 4, 2024 • 26