作者位次 #7/8ACL 2026Findings· Findings of ACL
a1: Steep Test-time Scaling Law via Environment Augmented Generation
Lingrui Mei · Shenghua Liu · Yiwei Wang · Baolong Bi · Yuyao Ge · Jun Wan · Yurong Wu · Xueqi Cheng
摘要
Large Language Models (LLMs) have made remarkable breakthroughs in reasoning, yet continue to struggle with hallucinations, logical errors, and inability to self-correct during complex multi-step tasks. Current approaches like chain-of-thought prompting offer limited reasoning capabilities that f...