作者位次 #21/25ACL 2026Poster· Natural Language Processing
A dvanced IF : Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Following
Yun He · Wenzhe Li · Hejia Zhang · Songlin Li · Karishma Mandyam · Sopan Khosla · Yuanhao Xiong · Nanshu Wang · Xiaoliang Peng · Beibin Li · Shengjie Bi · Shishir G. Patil · Qi Qi · Shengyu Feng · Julian Katz-Samuels · Richard Yuanzhe Pang · Sujan Kumar Gonugondla · Hunter Lang · Yue Yu · Yundi Qian · Maryam Fazel-Zarandi · Licheng Yu · Amine Benhalloum · Hany Hassan Awadalla · Manaal Faruqui
摘要
Recent progress in large language models (LLMs) has led to impressive performance on a range of tasks, yet advanced instruction following (IF)—especially for complex, multi-turn, and system-prompted instructions—remains a significant challenge. Rigorous evaluation and effective training for such...