The RealHumanEval: Evaluating Large Language Models' Abilities to Support Programmers
This work addresses the need for better human-centric evaluation of large language models as programmer assistants, though it is incremental in refining evaluation methods rather than introducing new paradigms.
The study tackled the problem of whether gains on existing benchmarks or preferred LLM responses translate to programmer productivity when coding with LLMs, finding that improvements in benchmark performance lead to increased programmer productivity, but gaps in benchmark versus human performance are not proportional, and programmer preferences do not correlate with actual performance.
Evaluation of large language models for code has primarily relied on static benchmarks, including HumanEval (Chen et al., 2021), or more recently using human preferences of LLM responses. As LLMs are increasingly used as programmer assistants, we study whether gains on existing benchmarks or more preferred LLM responses translate to programmer productivity when coding with LLMs, including time spent coding. We introduce RealHumanEval, a web interface to measure the ability of LLMs to assist programmers, through either autocomplete or chat support. We conducted a user study (N=243) using RealHumanEval in which users interacted with seven LLMs of varying base model performance. Despite static benchmarks not incorporating humans-in-the-loop, we find that improvements in benchmark performance lead to increased programmer productivity; however gaps in benchmark versus human performance are not proportional -- a trend that holds across both forms of LLM support. In contrast, we find that programmer preferences do not correlate with their actual performance, motivating the need for better proxy signals. We open-source RealHumanEval to enable human-centric evaluation of new models and the study data to facilitate efforts to improve code models.