TECH·2 days agoTencent Open-Sources AngelSpec: A Unified Training Framework for MTP and Block-Parallel Speculative Decoding on Hy3 ModelsMarkTechPost
TECH·May 27, 2026Meet EAGLE 3.1: The Speculative Decoding Algorithm That Fixes Attention Drift in LLM InferenceMarkTechPost
TECH·May 2, 2026New NVIDIA Research Shows Speculative Decoding in NeMo RL Achieves 1.8× Rollout Generation Speedup at 8B and Projects 2.5× End-to-End Speedup at 235BMarkTechPost
