BenchMIRT reveals hidden dimensions behind LLM benchmarks
BenchMIRT reveals hidden dimensions behind LLM benchmarks BenchMIRT is a new method for auditing large language model benchmarks at the level of individual prompts, allowing researchers to see which capabilities drive each score. Benchmarks are…