Ai Evaluation
No compiled wiki article for this topic yet. Raw entries below are the source material — a wiki article can be generated on demand from /admin/triggers.
All entries on this topic (1)
Challenging Medical LLM Evaluations: Polish Exam Benchmark Reveals Overstated Competence in Standard MCQA
Current evaluations of medical LLMs using standard multiple-choice question answering (MCQA) may inflate performance due to guessing and answer biases. A newly developed benchmark, featuring over 15,000 questions from Polish medical exams with structural modifications to mitigate MCQA artifacts, rev…