Scientists' First Exam: Probing Cognitive Abilities of MLLM via Perception, Understanding, and Reasoning
This addresses the problem of inadequate assessment of MLLMs' cognitive abilities in scientific domains for researchers and developers, though it is incremental as it builds on existing benchmarking efforts.
The authors tackled the lack of comprehensive evaluation for scientific multimodal large language models (MLLMs) by introducing the Scientists' First Exam (SFE) benchmark, which assesses perception, understanding, and reasoning across 66 tasks, revealing that top models like GPT-o3 and InternVL-3 achieve only 34.08% and 26.52% accuracy.
Scientific discoveries increasingly rely on complex multimodal reasoning based on information-intensive scientific data and domain-specific expertise. Empowered by expert-level scientific benchmarks, scientific Multimodal Large Language Models (MLLMs) hold the potential to significantly enhance this discovery process in realistic workflows. However, current scientific benchmarks mostly focus on evaluating the knowledge understanding capabilities of MLLMs, leading to an inadequate assessment of their perception and reasoning abilities. To address this gap, we present the Scientists' First Exam (SFE) benchmark, designed to evaluate the scientific cognitive capacities of MLLMs through three interconnected levels: scientific signal perception, scientific attribute understanding, scientific comparative reasoning. Specifically, SFE comprises 830 expert-verified VQA pairs across three question types, spanning 66 multimodal tasks across five high-value disciplines. Extensive experiments reveal that current state-of-the-art GPT-o3 and InternVL-3 achieve only 34.08% and 26.52% on SFE, highlighting significant room for MLLMs to improve in scientific realms. We hope the insights obtained in SFE will facilitate further developments in AI-enhanced scientific discoveries.