ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Zhilin, Jung, Jaehun, Lu, Ximing, Diao, Shizhe, Evans, Ellie, Zeng, Jiaqi, Molchanov, Pavlo, Choi, Yejin, Kautz, Jan, Dong, Yi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916019050643456
author Wang, Zhilin
Jung, Jaehun
Lu, Ximing
Diao, Shizhe
Evans, Ellie
Zeng, Jiaqi
Molchanov, Pavlo
Choi, Yejin
Kautz, Jan
Dong, Yi
author_facet Wang, Zhilin
Jung, Jaehun
Lu, Ximing
Diao, Shizhe
Evans, Ellie
Zeng, Jiaqi
Molchanov, Pavlo
Choi, Yejin
Kautz, Jan
Dong, Yi
contents Evaluating progress in large language models (LLMs) is often constrained by the challenge of verifying responses, limiting assessments to tasks like mathematics, programming, and short-form question-answering. However, many real-world applications require evaluating LLMs in processing professional documents, synthesizing information, and generating comprehensive reports in response to user queries. We introduce ProfBench: a set of over 7000 response-criterion pairs as evaluated by human-experts with professional knowledge across Physics PhD, Chemistry PhD, Finance MBA and Consulting MBA. We build robust and affordable LLM-Judges to evaluate ProfBench rubrics, by mitigating self-enhancement bias and reducing the cost of evaluation by 2-3 orders of magnitude, to make it fair and accessible to the broader community. Our findings reveal that ProfBench poses significant challenges even for state-of-the-art LLMs, with top-performing models like GPT-5-high achieving only 65.9% overall performance. Furthermore, we identify notable performance disparities between proprietary and open-weight models and provide insights into the role that extended thinking plays in addressing complex, professional-domain tasks. Data: https://huggingface.co/datasets/nvidia/ProfBench and Code: https://github.com/NVlabs/ProfBench and Leaderboard: https://huggingface.co/spaces/nvidia/ProfBench
format Preprint
id arxiv_https___arxiv_org_abs_2510_18941
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge
Wang, Zhilin
Jung, Jaehun
Lu, Ximing
Diao, Shizhe
Evans, Ellie
Zeng, Jiaqi
Molchanov, Pavlo
Choi, Yejin
Kautz, Jan
Dong, Yi
Computation and Language
Artificial Intelligence
Machine Learning
Evaluating progress in large language models (LLMs) is often constrained by the challenge of verifying responses, limiting assessments to tasks like mathematics, programming, and short-form question-answering. However, many real-world applications require evaluating LLMs in processing professional documents, synthesizing information, and generating comprehensive reports in response to user queries. We introduce ProfBench: a set of over 7000 response-criterion pairs as evaluated by human-experts with professional knowledge across Physics PhD, Chemistry PhD, Finance MBA and Consulting MBA. We build robust and affordable LLM-Judges to evaluate ProfBench rubrics, by mitigating self-enhancement bias and reducing the cost of evaluation by 2-3 orders of magnitude, to make it fair and accessible to the broader community. Our findings reveal that ProfBench poses significant challenges even for state-of-the-art LLMs, with top-performing models like GPT-5-high achieving only 65.9% overall performance. Furthermore, we identify notable performance disparities between proprietary and open-weight models and provide insights into the role that extended thinking plays in addressing complex, professional-domain tasks. Data: https://huggingface.co/datasets/nvidia/ProfBench and Code: https://github.com/NVlabs/ProfBench and Leaderboard: https://huggingface.co/spaces/nvidia/ProfBench
title ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2510.18941