Unverified50% confidenceFactExact time
HumanEval通过pass@k指标衡量模型生成正确函数实现的概率
1
Sources
50%
Confidence
Long-term
Relevance
7/12/2026
First Seen
Sources
GPT-5来了，自学Java还有用吗？AI时代程序员如何转型
bilibiliAI开发旗博士7/9/2026
Related Claims
Unverified代码模型采用HumanEval、MBPP等可执行基准进行评估，通用语言模型则依赖困惑度等统计指标63% similarUnverified模型选型时应以每美元token产出或每瓦特推理性能替代单一准确率指标进行评估59% similarUnverifiedBayesian optimization constructs a probabilistic surrogate model of the objective function (typically a Gaussian process) and uses an acquisition function (such as Expected Improvement or Upper Confidence Bound) to determine the next sampling point.59% similarUnverified贝叶斯优化基于概率代理模型（通常为高斯过程）建模历史实验结果，在探索与利用之间智能权衡，通常能以更少实验次数找到接近最优的参数配置56% similarUnverified贝叶斯知识追踪（BKT）模型将每个知识点建模为二值隐状态，通过隐马尔可夫模型估计'已掌握'概率56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/489201API
curl https://kongchang.com/api/v1/knowledge/claims/489201MCP
get_claim(id=489201)