None of the Others: a General Technique to Distinguish Reasoning from Memorization in Multiple-Choice LLM Evaluation Benchmarks

Documenti analoghi

• Simulating Training Data Leakage in Multiple-Choice Benchmarks for LLM Evaluation
  di: Hidayat, Naila Shafirni, et al.
  Pubblicazione: (2025)
• Bilingual Evaluation of Language Models on General Knowledge in University Entrance Exams with Minimal Contamination
  di: Salido, Eva Sánchez, et al.
  Pubblicazione: (2024)
• Multiple-Choice Questions are Efficient and Robust LLM Evaluators
  di: Zhang, Ziyin, et al.
  Pubblicazione: (2024)
• Beyond Memorization: Distinguishing between Reductive and Epistemic Reasoning in LLMs using Classic Logic Puzzles
  di: Gabay, Adi, et al.
  Pubblicazione: (2026)
• The Reader is the Metric: How Textual Features and Reader Profiles Explain Conflicting Evaluations of AI Creative Writing
  di: Marco, Guillermo, et al.
  Pubblicazione: (2025)