EyePCR: A Comprehensive Benchmark for Fine-Grained Perception, Knowledge Comprehension and Clinical Reasoning in Ophthalmic Surgery

MLLMs (Multimodal Large Language Models) have showcased remarkable capabilities, but their performance in high-stakes, domain-specific scenarios like surgical settings, remains largely under-explored. To address this gap, we develop $\mathbf{\text{EyePCR}}$, a large-scale benchmark for ophthalmic surgery analysis, grounded in structured clinical knowledge to evaluate cognition across $\mathit{\text{Perception}}$, $\mathit{\text{Comprehension}}$ and $\mathit{\text{Reasoning}}$. EyePCR offers a richly annotated corpus with more than 210k VQAs, which cover 1048 fine-grained attributes for multi-view perception, medical knowledge graph of more than 25k triplets for comprehension, and four clinically grounded reasoning tasks. The rich annotations facilitate in-depth cognitive analysis, simulating how surgeons perceive visual cues and combine them with domain knowledge to make decisions, thus greatly improving models’ cognitive ability. In particular, $\mathbf{\text{EyePCR-MLLM}}$, a domain-adapted variant of Qwen2.5-VL-7B, achieves the highest accuracy on MCQs for $\mathit{\text{Perception}}$ among compared models and outperforms open-source models in $\mathit{\text{Comprehension}}$ and $\mathit{\text{Reasoning}}$, rivalling commercial models like GPT-4.1. EyePCR reveals the limitations of existing MLLMs in surgical cognition and lays the foundation for benchmarking and enhancing clinical reliability of surgical video understanding models.