作者位次 #3/6ACL 2026Poster· Natural Language Processing
A gent C o M a: A Compositional Benchmark Mixing Commonsense and Mathematical Reasoning in Real-World Scenarios
Lisa Alazraki · Lihu Chen · Ana Brassard · Joe Stacey · Hossein A. Rahmani · Marek Rei
摘要
Large Language Models (LLMs) have achieved high accuracy on complex commonsense and mathematical problems that involve the composition of multiple reasoning steps. However, current compositional benchmarks testing these skills tend to focus on either commonsense or math reasoning, whereas LLM age...