ResearchBenchmarksMulti-label topic classification on DVD playerFollow69.9Micro-average F1AK49.30854.6546065.346May 28, 2026Evaluation ResultsMethodMethodLinksMicro-average F1AKModel=GPT-4o, Graph-fr...Model=GPT-4o, Graph-free=true2026.0569.9AKSModel=Qwen 2.5-32B, Gr...Model=Qwen 2.5-32B, Graph-free=true2026.0569.9BaselineSource=Sarkar et al.,...Source=Sarkar et al., 20232026.0550.1