Unverified50% confidenceFactExact time
Frontier Code的误报率为6.9%,相比SWE-Bench Pro的36.0%降低了约81%
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Frontier Code深度解析:代码能跑≠能合并,编程基准测试迎来质量革命
bilibili烟神殿副殿主6/9/2026
Related Claims
UnverifiedBackEngine 团队宣称相比直连连接器,错误率降低 67%67% similarUnverifiedSWE-Bench Pro的假阳性率为8.5%,假阴性率高达24%;而DeepSWE的假阳性率仅为0.3%,假阴性率仅为1.1%66% similarUnverified单纯使用'少写代码'提示词可减少33%代码量,但安全指标下降至95%66% similarUnverifiedSWE-Bench Pro has a false negative rate of 24%, meaning 1 out of every 4 correct answers is incorrectly marked as a failure.66% similarUnverifiedSWE-Bench Pro has a false positive rate of 8.5%, meaning the verifier accepts incorrect implementations at that rate.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/49041API
curl https://kongchang.com/api/v1/knowledge/claims/49041MCP
get_claim(id=49041)