Human Eval
2 mentions across 2 people
Visit ↗All mentions
“my team collected some data from a coding Benchmark um the benchmark's name is Human eval but it just evaluates how well an AI system is able to write code to solve the low coding puzzles”
Emerging AI Agent Design Patterns Drive Application Layer Innovation and Corpora ↗
“at the very beginning of the project we wrote down this data set that's now open source called we call it human eval which is a list of problems written by humans that are just programming puzzles”
OpenAI’s Codex: A GPT-3 Descendant for Code Generation ↗