Fast and Exact Relative-Position Attention via Structure-Aware Kernels Paper • 2611.01182 • Published 14 days ago • 23
Why Pay for Long Reasoning When Rewrite Do Trick? Causal Effects of Reasoning History Compaction on Future Generation and Agent Cost Paper • 2611.03852 • Published 12 days ago • 18
When Does Latent Prediction Error Support Physical Surprise Claims in Video World Models? Paper • 2611.01815 • Published 12 days ago • 29