benchmark designed to evaluate large language models (LLMs) on solving complex, college-level scientific problems from domains like chemistry, physics, and mathematics.

evaluates LLM's ability to call external functions/tools.

a meta-benchmark that evaluates how well factuality evaluators assess the outputs of large language models (LLMs).

a benchmark that evaluates large language models on a variety of multimodal reasoning tasks, including language, natural and social sciences, physical and social commonsense, temporal reasoning, algebra, and geometry.

focuses on understanding how these models perform in various scenarios and analyzing results from an interpretability perspective.

a benchmark designed to assess the performance of multimodal web agents on realistic visually grounded tasks.