GPT-5 with source-citation prompting achieves 89.1% accuracy on 993 PSM questions, outperforming zero-shot and chain-of-thought while errors cluster in multi-select and interpretive topics.
Machine vs machine: Large language models (llms) in applied machine learning high-stakes open-book exams,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Prompting GPT-5 on Scrum Certification Questions: An Empirical Accuracy Study
GPT-5 with source-citation prompting achieves 89.1% accuracy on 993 PSM questions, outperforming zero-shot and chain-of-thought while errors cluster in multi-select and interpretive topics.