AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

How Much Do Large Language Model Cheat on Evaluation? Benchmarking Overestimation Under the One-Time-Pad-Based Framework

Zi Liang, Liantong Yu, Zhang Shiyu, Qingqing Ye, Haibo Hu

AAAI 2026 · Large language models › Language, safety and interpretability · approach: Uncertainty quantification for LLMs · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
regenerate SCP test cases with 32 different random seeds from the same articles (ablation)ArxivRollBench-2024b-CS · ArxivRollBench-2024b-CS, Llama3-8Baccuracy std: std below 1 point (0.0074 to 0.0095)No clear effect
use Sequencing vs Prediction vs Cloze generation strategy (ablation)ArxivRollBench · Correlation with ChatbotArena (Spearman)Spearman correlation: S 0.76, P 0.73, C 0.61Mixed
newer model versions within a series (e.g., Phi series)LLM series · ArxivRollBench vs public benchmarksaccuracy and RSIMixed
measure public vs private benchmark gapQwen and Phi series · public vs ArxivRollBenchAbsolute RSI: Qwen2.5-72B 1.41; Phi-3-mini 1.27Hurt
compare public vs private benchmark performance by domainGPT-J-6B, Phi2, Llama3-8B, Yi1.5-34B, Llama3.1-Nemotron-70B · Econ, Q-Fin, Bio, Phy domainsAbsolute RSIHurt
scale up to Llama3.1-70B or Nemotron-70BLlama family · Private benchmarksRS_II vs normalized RS_II: Nemotron-70B RS_II 1.30%, normalized 3.88%Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy