← back to paper

Review history

arxiv: 2604.11530 · 2 revisions

Beyond Attention Scores: SVD-Based Vision Token Pruning for Efficient Vision-Language Models

1. 2026-05-21 UNVERDICTED LOW v0.9.0 novelty 6.0
   38304 ms 5695 in 1176 out 2026-05-21T08:35:59.498292+00:00
2. 2026-05-10 UNVERDICTED LOW v0.9.0 novelty 5.0
   33976 ms 5462 in 1151 out 2026-05-10T16:22:57.618953+00:00