NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmark

  1. Sainz, O.
  2. Campos, J.A.
  3. García-Ferrero, I.
  4. Etxaniz, J.
  5. Lopez de Lacalle, O.
  6. Agirre, E.
Proceedings:
Findings of the Association for Computational Linguistics: EMNLP 2023

ISBN: 9798891760615

Year of publication: 2023

Findings of the Association for Computational Linguistics: EMNLP 2023

Pages: 10776-10787

Type: Conference paper