JetBrains-Research/sft-router-qwen3-4b-swe-bench Text Generation โข 4B โข Updated 5 days ago โข 45
JetBrains-Research/sft-router-qwen3-4b-swe-bench Text Generation โข 4B โข Updated 5 days ago โข 45
LK Losses: Direct Acceptance Rate Optimization for Speculative Decoding Paper โข 2602.23881 โข Published 26 days ago โข 18