HippoMM: Hippocampal-inspired Multimodal Memory for Long Audiovisual Event Understanding Paper • 2504.10739 • Published Apr 14 • 2
CoreMatching: A Co-adaptive Sparse Inference Framework with Token and Neuron Pruning for Comprehensive Acceleration of Vision-Language Models Paper • 2505.19235 • Published May 25 • 3
Angles Don't Lie: Unlocking Training-Efficient RL Through the Model's Own Signals Paper • 2506.02281 • Published Jun 2 • 4
Voice Evaluation of Reasoning Ability: Diagnosing the Modality-Induced Performance Gap Paper • 2509.26542 • Published Sep 30 • 8