Picture for Huichuan Zheng

Huichuan Zheng

SPORK: Self-Speculative Forking to Accelerate Agentic LLM Inference

Add code
Jul 03, 2026
Viaarxiv icon