Unverified50% confidenceOpinionExact time
Evaluations based on open-source codebases suffer from data contamination issues, as these codebases may have been included in large model training data.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe training data sources for specialized programming models include open-source code from GitHub, competitive programming solutions, technical documentation, and code review records, and the models are optimized for benchmarks such as HumanEval, MBPP, and SWE-bench.79% similarVerifiedTraining data contamination is the primary challenge faced by AI programming benchmarks, as test cases come from public GitHub repositories, which are the same sources used as pretraining data for large models.79% similarUnverified数据污染问题源于LLM预训练语料来自Common Crawl等大规模网络爬取，可能包含公开的基准测试题目及答案75% similarUnverifiedForms of information leakage in code evaluation include test case visibility leakage, environment metadata leakage, and training data contamination.75% similarUnverifiedLLM code auditing reasons based on code patterns and known vulnerability samples learned during training, making it difficult to perform precise state tracking across files and cross-function call chains.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/917814API
curl https://kongchang.com/api/v1/knowledge/claims/917814MCP
get_claim(id=917814)