Category Latest News

Multimodal Foundation Models Fall Short on Physical Reasoning: PHYX Benchmark Highlights Key Limitations in Visual and Symbolic Integration

State-of-the-art models show human-competitive accuracy on AIME, GPQA, MATH-500, and OlympiadBench, solving Olympiad-level problems. Recent multimodal foundation models have advanced benchmarks for disciplinary knowledge and mathematical reasoning. However, these evaluations miss a crucial aspect of machine intelligence: physical reasoning, which…