mlsanityData to 14 Oct 2026

Towards Contamination Resistant Benchmarks

arXiv 2505.08389 ↗First posted 13 May 2025cs.CLcited by 0

Not on the map. Why?

Abstract (arXiv)

The rapid development of large language models (LLMs) has transformed the landscape of natural language processing. Evaluating LLMs properly is crucial for understanding their potential and addressing concerns such as safety. However, LLM evaluation is confronted by various factors, among which contamination stands out as a key issue that undermines the reliability of evaluations. In this work, we introduce the concept of contamination resistance to address this challenge. We propose a benchmark based on Caesar ciphers (e.g., "ab" to "bc" when the shift is 1), which, despite its simplicity, is an excellent example of a contamination resistant benchmark. We test this benchmark on widely used LLMs under various settings, and we find that these models struggle with this benchmark when contamination is controlled. Our findings reveal issues in current LLMs and raise important questions regarding their true capabilities. Our work contributes to the development of contamination resistant benchmarks, enabling more rigorous LLM evaluation and offering insights into the true capabilities and limitations of LLMs.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 44 of its 56 references that are in the map. First posted on arXiv 13 May 2025.

Builds on

44 of 56 references.

References · 44

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  3. 2021Training Verifiers to Solve Math Word Problemscited by 10,109
  4. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,326
  5. 2024Qwen2.5 Technical Reportcited by 4,942
  6. 2022Large Language Models are Zero-Shot Reasonerscited by 4,374
  7. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  8. 2023Sparks of Artificial General Intelligence: Early experiments with GPT-4cited by 2,592
  9. 2022Emergent Abilities of Large Language Modelscited by 2,512
  10. 2024LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Codecited by 2,004
  11. 2022Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language modelscited by 1,998
  12. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
32 more
  1. 2019SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systemscited by 1,729
  2. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,554
  3. 2023A Multitask, Multilingual, Multimodal Evaluation of ChatGPT on Reasoning, Hallucination, and Interactivitycited by 894
  4. 2022What Can Transformers Learn In-Context? A Case Study of Simple Function Classescited by 715
  5. 2022What learning algorithm is in-context learning? Investigations with linear modelscited by 684
  6. 2021Prompt Programming for Large Language Models: Beyond the Few-Shot Paradigmcited by 606
  7. 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 530
  8. 2023Are Emergent Abilities of Large Language Models a Mirage?cited by 523
  9. 2023NLP Evaluation in trouble: On the Need to Measure LLM Data Contamination for each Benchmarkcited by 348
  10. 2023Adaptive Chameleon or Stubborn Sloth: Revealing the Behavior of Large Language Models in Knowledge Conflictscited by 341
  11. 2022Language Models Are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thoughtcited by 337
  12. 2023Trained Transformers Learn Linear Models In-Contextcited by 337
  13. 2023How Language Model Hallucinations Can Snowballcited by 311
  14. 2022Emergent Analogical Reasoning in Large Language Modelscited by 240
  15. 2024Leak, Cheat, Repeat: Data Contamination and Evaluation Malpractices in Closed-Source LLMscited by 229
  16. 2024A Careful Examination of Large Language Model Performance on Grade School Arithmeticcited by 191
  17. 2023Sources of Hallucination by Large Language Models on Inference Taskscited by 169
  18. 2022Thinking Fast and Slow in Large Language Modelscited by 137
  19. 2023Self-contradictory Hallucinations of Large Language Models: Evaluation, Detection and Mitigationcited by 124
  20. 2024Generalization or Memorization: Data Contamination and Trustworthy Evaluation for Large Language Modelscited by 122
  21. 2023Are Emergent Abilities in Large Language Models just In-Context Learning?cited by 109
  22. 2023Task Contamination: Language Models May Not Be Few-Shot Anymorecited by 99
  23. 2023Language Models Implement Simple Word2Vec-style Vector Arithmeticcited by 98
  24. 2023Instructed to Bias: Instruction-Tuned Language Models Exhibit Emergent Cognitive Biascited by 53
  25. 2023How are Prompts Different in Terms of Sensitivity?cited by 35
  26. 2025Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluationcited by 29
  27. 2024TRUCE: Private Benchmarking to Prevent Contamination and Improve Comparative Evaluation of LLMscited by 24
  28. 2023In-context Learning Generalizes, But Not Always Robustly: The Case of Syntaxcited by 22
  29. 2024When a language model is optimized for reasoning, does it still show embers of autoregression? An analysis of OpenAI o1cited by 22
  30. 2024LiveXiv -- A Multi-Modal Live Benchmark Based on Arxiv Papers Contentcited by 17
  31. 2022WinoDict: Probing language models for in-context word acquisitioncited by 15
  32. 2022Can In-context Learners Learn a Reasoning Concept from Demonstrations?cited by 7

Built on by

Not cited yet.