Login / Signup

Evading Data Contamination Detection for Language Models is (too) Easy.

Jasper DekoninckMark Niklas MüllerMaximilian BaaderMarc FischerMartin T. Vechev
Published in: CoRR (2024)
Keyphrases
  • language model
  • information retrieval
  • n gram
  • machine learning
  • training data
  • query processing
  • language modeling
  • probabilistic model
  • information extraction
  • speech recognition
  • vector space
  • context sensitive