C2LEVA: Toward Comprehensive and Contamination-Free Language Model Evaluation
Not on the map. Why?
Abstract (arXiv)
Recent advances in large language models (LLMs) have shown significant promise, yet their evaluation raises concerns, particularly regarding data contamination due to the lack of access to proprietary training data. To address this issue, we present C2LEVA, a comprehensive bilingual benchmark featuring systematic contamination prevention. C2LEVA firstly offers a holistic evaluation encompassing 22 tasks, each targeting a specific application or ability of LLMs, and secondly a trustworthy assessment due to our contamination-free tasks, ensured by a systematic contamination prevention strategy that fully automates test data renewal and enforces data protection during benchmark data release. Our large-scale evaluation of 15 open-source and proprietary models demonstrates the effectiveness of C2LEVA.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 30 of its 52 references that are in the map. First posted on arXiv 6 Dec 2024.
Builds on
30 of 52 references.
References · 30
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
- 2021Training Verifiers to Solve Math Word Problemscited by 10,109
- 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
- 2020Measuring Massive Multitask Language Understandingcited by 7,963
- 2024Gemini 1.5: Unlocking multimodal understanding across millions of tokens of contextcited by 3,224
- 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
- 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
- 2020Extracting Training Data from Large Language Modelscited by 2,189
- 2019Language Models as Knowledge Bases?cited by 1,936
- 2016Enriching Word Vectors with Subword Informationcited by 1,583
- 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
18 more
- 2023Large Language Models Can Be Easily Distracted by Irrelevant Contextcited by 841
- 2019CCNet: Extracting High Quality Monolingual Datasets from Web Crawl Datacited by 548
- 2021Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpuscited by 492
- 2021True Few-Shot Learning with Language Modelscited by 344
- 2023Proving Test Set Contamination in Black Box Language Modelscited by 230
- 2023Rethinking Benchmark and Contamination for Language Models with Rephrased Samplescited by 195
- 2023Don't Make Your LLM an Evaluation Benchmark Cheatercited by 194
- 2023Stop Uploading Test Data in Plain Text: Practical Strategies for Mitigating Data Contamination by Evaluation Benchmarkscited by 142
- 2023Skywork: A More Open Bilingual Foundation Modelcited by 112
- 2023NPHardEval: Dynamic Benchmark on Reasoning Ability of Large Language Models via Complexity Classescited by 79
- 2024Benchmark Self-Evolving: A Multi-Agent Framework for Dynamic LLM Evaluationcited by 75
- 2024DE-COP: Detecting Copyrighted Content in Language Models Training Datacited by 74
- 2024MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixturescited by 74
- 2023LatestEval: Addressing Data Contamination in Language Model Evaluation through Dynamic and Time-Sensitive Test Constructioncited by 58
- 2024Pretraining Data Detection for Large Language Models: A Divergence-based Calibration Methodcited by 55
- 2024Proving membership in LLM pretraining data via data watermarkscited by 40
- 2024VarBench: Robust Language Model Benchmarking Through Dynamic Variable Perturbationcited by 29
- 2023CLEVA: Chinese Language Models EVAluation Platformcited by 13
Built on by
Citing papers per year: 2025: 2.
2 papers.
- 2025NeMo: Needle in a Montage for Video-Language Understanding29 Sep 2025
- 2025SCAN: Structured Capability Assessment and Navigation for LLMs10 May 2025