Unverified50% confidenceFactExact time
Alpha Codium research shows that a closed-loop approach of "plan—write—run tests—see errors—fix—repeat" raises GPT-4's pass rate on the CodeContests benchmark from 19% to 44%.
1
Sources
50%
Confidence
Long-term
Relevance
10/18/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedWhen checking code against written specifications, GPT-4o falsely failed 26.2% and 35.9% of correct programs on two benchmarks respectively, and if it was also required to explain and fix the code, the false-failure rate soared to 73.2% and 87.9%.75% similarUnverifiedCarlini等人2023年的研究显示，GPT-4等模型可以在特定提示下逐字复现训练集中的代码片段74% similarUnverifiedA 2023 study found that GPT-4 produced errors in approximately 15–20% of cases when generating code documentation.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1059518API
curl https://kongchang.com/api/v1/knowledge/claims/1059518MCP
get_claim(id=1059518)