IntBMoE: Integrating Block-Level Conditioning into Expert Composition for Full-Participation Mixture-of-Experts Paper • 2609.21346 • Published 5 days ago • 94
Full Attention Strikes Back: Transferring Full Attention into Sparse within Hundred Training Steps Paper • 2605.16928 • Published May 16 • 91
TransitLM: A Large-Scale Dataset and Benchmark for Map-Free Transit Route Generation Paper • 2605.22355 • Published May 21 • 128