HumanEvalNext

This is the dataset for the paper https://arxiv.org/abs/2503.05860 HumanEvalNext HumanEvalNext is an improved version of the HumanEval code generation benchmark. The improvements made are based on the framework EvalPro. This framework is aimed at enhancing benchmark quality through a rigorous improvement process along

Papers using HumanEvalNext (2)

Benchmarking AI Models in Software Engineering: A Review, Search Tool, and Unified Approach for Elevating Benchmark Quality2025 · 1 citesBenchmarking AI Models in Software Engineering: A Review, Search Tool, and Enhancement Protocol2025