作者位次 #17/24ACL 2026Poster· Natural Language Processing
PRB ench: Large-Scale Expert Rubrics for Evaluating High-Stakes Professional Reasoning
Afra Feyza Akyürek · Advait Gosai · Chen Bo Calvin Zhang · Vipul Gupta · Jaehwan Jeong · Anisha Gunjal · Tahseen Rabbani · Maria Mazzone · David Randolph IV · Mohammad Mahmoudi Meymand · Gurshaan Chattha · Paula Rodriguez · Diego A. Mares Buendia · Pavit Singh · Michael Liu · Subodh Chawla · Peter Cline · Lucy Ogaz · Ernesto Gabriel Hernández Montoya · Zihao Wang · Pavi Bhatter · Marcos Ayestaran · Bing Liu · Yunzhong He
摘要
Frontier model progress is often measured using academic benchmarks that provide a limited view of performance on open-ended, economically consequential tasks in high-stakes professional domains where practical returns matter most. We introduce Professional Reasoning Bench (PRBench), a realistic,...