作者位次 #2/7ACL 2026Findings· Findings of ACL
G limp R outer: Efficient Collaborative Inference by Glimpsing One Token of Thoughts
Wenhao Zeng · Xuteng Zhang · Yuling Shi · Chao Hu · Yuting Chen · Beijun Shen · Xiaodong Gu
摘要
Large Reasoning Models (LRMs) achieve remarkable performance by explicitly generating multi-step chains of thought, but this capability incurs substantial inference latency and computational cost. Collaborative inference offers a promising solution by selectively allocating work between lightweig...