Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Zero Gap Is Not Restoration: Stratified Per-Question Probability Evaluation and Step-wise Mitigation of Benchmark Contamination
Ruijie Hou, Yueyang Jiao, Zhao Wang +1
Test data from public benchmarks inevitably leaks into pretraining corpora, inflating evaluation scores once memorized. \textbf{Contamination mitigation evaluation} intervenes in t…
cs.CL2025
LNE-Blocking: An Efficient Framework for Contamination Mitigation Evaluation on Large Language Models
Ruijie Hou, Yueyang Jiao, Hanxu Hu +4
The problem of data contamination is now almost inevitable during the development of large language models (LLMs), with the training data commonly integrating those evaluation benc…