作者位次 #3/4ACL 2026Findings· Findings of ACL
Better LLM Reasoning via Dual-Play
Zhengxin Zhang · Chengyu Huang · Aochong Oliver Li · Claire Cardie
摘要
Large Language Models (LLMs) have achieved remarkable progress through Reinforcement Learning with Verifiable Rewards (RLVR), yet still rely heavily on external supervision (e.g., curated labels). Adversarial learning, particularly through self-play, offers a promising alternative that enables mo...