ProBench: Judging Multimodal Foundation Models on Open-ended Multi-domain Expert Tasks
This work addresses the need for better evaluation of multimodal AI models for professionals, though it is incremental as it builds on existing benchmarking approaches.
The authors tackled the challenge of evaluating multimodal large language models on expert-level tasks by introducing ProBench, a benchmark of 4,000 open-ended queries across 10 fields, and found that while top open-source models compete with proprietary ones, significant gaps remain in visual perception, textual understanding, domain knowledge, and advanced reasoning.
Solving expert-level multimodal tasks is a key milestone towards general intelligence. As the capabilities of multimodal large language models (MLLMs) continue to improve, evaluation of such advanced multimodal intelligence becomes necessary yet challenging. In this work, we introduce ProBench, a benchmark of open-ended user queries that require professional expertise and advanced reasoning. ProBench consists of 4,000 high-quality samples independently submitted by professionals based on their daily productivity demands. It spans across 10 fields and 56 sub-fields, including science, arts, humanities, coding, mathematics, and creative writing. Experimentally, we evaluate and compare 24 latest models using MLLM-as-a-Judge. Our results reveal that although the best open-source models rival the proprietary ones, ProBench presents significant challenges in visual perception, textual understanding, domain knowledge and advanced reasoning, thus providing valuable directions for future multimodal AI research efforts.