Unverified50% confidenceFactExact time
根据对NeurIPS、ICML等顶级会议论文代码的调查，约60%的开源研究代码缺乏完整的依赖说明，超过70%仅在特定Linux发行版上测试过
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/26/2026
First Seen
Valid until: 11/24/2026
Sources
Related Entities
Related Claims
UnverifiedAccording to Sonar research, 96% of people do not fully trust AI-generated code, yet only about half always verify it before submission.66% similarUnverified人工代码审查平均只能发现约60%的缺陷64% similarUnverifiedResearch data from the Standish Group shows that approximately 64% of features in software projects are rarely or never used.63% similarUnverified机器学习领域约有30%-50%的论文没有公开可用的代码63% similarUnverifiedA developer cited in the article found that more than 95% of the code provided by AI was accepted directly, with only 5% actually requiring manual modification.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/802026API
curl https://kongchang.com/api/v1/knowledge/claims/802026MCP
get_claim(id=802026)