BenchMarker: An Education-Inspired Toolkit for Highlighting Flaws in Multiple-Choice Benchmarks
This work addresses benchmark reliability issues in NLP evaluation, offering a practical tool for researchers, though it is incremental in applying education-inspired methods to existing problems.
The authors tackled the problem of quality flaws in multiple-choice question answering benchmarks by developing BenchMarker, a toolkit that uses LLM judges to detect contamination, shortcuts, and writing errors; validation on 12 benchmarks revealed that contaminated items inflate accuracy while writing errors lower it and affect rankings.
Multiple-choice question answering (MCQA) is standard in NLP, but benchmarks lack rigorous quality control. We present BenchMarker, an education-inspired toolkit using LLM judges to flag three common MCQ flaws: 1) contamination - items appearing exactly online; 2) shortcuts - cues in the choices that enable guessing; and 3) writing errors - structural/grammatical issues based on a 19-rule education rubric. We validate BenchMarker with human annotations, then run the tool to audit 12 benchmarks, revealing: 2) contaminated MCQs tend to inflate accuracy, while writing errors tend to lower it and change rankings beyond random; and 3) prior benchmark repairs address their targeted issues (i.e., lowering accuracy with LLM-written distractors), but inadvertently add new flaws (i.e. implausible distractors, many correct answers). Overall, flaws in MCQs degrade NLP evaluation, but education research offers a path forward. We release BenchMarker to bridge the fields and improve MCQA benchmark design.