Impulse2000/simple_bench_public-20-12-2024

Simple Bench Public Where Everyday Human Reasoning Still Surpasses Frontier Models. Dataset Details Dataset Description "[...] A multiple-choice text benchmark for LLMs where individuals with unspecialized (high school) knowledge outperform SOTA models. SimpleBench includes over 200 questions covering spatio-temporal reasoning, social intelligence, and what we call linguistic adversarial robustness (or trick questions). For the vast majority of… See the full description on the dataset page: https://huggingface.co/datasets/Impulse2000/simple_bench_public-20-12-2024.