Is ALMANACS a Simulatability Benchmark for Language Model Explainability?

Original title: ALMANACS: A Simulatability Benchmark for Language Model Explainability Authors: Edmund Mills, Shiye Su, Stuart Russell, Scott Emmons In this article, the authors discuss the challenge of measuring the effectiveness of language model explainability…

Read more of Is ALMANACS a Simulatability Benchmark for Language Model Explainability?