作者位次 #10/10ACL 2026Poster· Natural Language Processing
Knowledge-to-Verification: Exploring RLVR for LLM s in Knowledge-Intensive Domains
Zhonghang Yuan · Zhefan Wang · Fang Hu · Zihong Chen · Jinzhe Li · Gang Li · Jie Ying · Huanjun Kong · Songyang Zhang · Nanqing Dong
摘要
Reinforcement learning with verifiable rewards (RLVR) has demonstrated promising potential to enhance the reasoning capabilities of large language models (LLMs) in domains such as mathematics and coding. However, its applications on knowledge-intensive domains have not been effectively explored d...