AI RESEARCH
EventPrune: Cascaded Event-Assisted Token Pruning for Efficient First-Person Dynamic Spatial Reasoning
arXiv CS.CV
•
ArXi:2605.19506v1 Announce Type: new First-person dynamic spatial reasoning requires models to track continuous motion and precise geometric structure, but the quadratic attention cost of Transformer-based Video-LLMs makes dense visual tokens computationally expensive. Existing token pruning paradigms predominantly rely on discrete static snapshots, failing to preserve the motion and geometric cues essential for reasoning. We propose Event Cascade Pruning (ECP), to our knowledge the first.