The Posterior

A blog about learning machine learning, AI and statistics, all the way to the core ideas.

  • Explainers on ML, AI and statistics
  • Paper reviews and overviews
  • Now and then, opinions on tech and AI (everyone has some)
  • Once in a while, advice on research and life

Latest

  1. Post

    How well do JEV models know when they are right?

    When a decision model says it's 97% sure, how often is it right? We checked eight open models against real answers. On average, beautifully. Up close, it depends on the question.

  2. Leaderboard

    Decision benchmarks leaderboard

    Accuracy and calibration of Jev, other hosted decision models and open-weight ones on the decision benchmarks their makers use, broken down by multiple-choice, yes/no and rating questions.

  3. Leaderboard

    JEV calibration leaderboard

    Whether Jev, thirteen other hosted decision models and eight open-weight ones are right as often as they say they are, all asked the same exam-style questions.