AI RESEARCH
ReactBench: A Benchmark for Topological Reasoning in MLLMs on Chemical Reaction Diagrams
arXiv CS.AI
•
ArXi:2604.15994v1 Announce Type: new Multimodal Large Language Models (MLLMs) excel at recognizing individual visual elements and reasoning over simple linear diagrams. However, when faced with complex topological structures involving branching paths, converging flows, and cyclic dependencies, their reasoning capabilities degrade sharply, even on tasks as basic as counting endpoints. Existing benchmarks fail to probe this gap, focusing on semantic comprehension rather than structural reasoning. We