Themisto: Jupyter-Based Runtime Benchmark

Abstract

In this work, we present a benchmark that consists of Jupyter notebooks development trajectories and allows measuring how large language models (LLMs) can leverage runtime information for predicting code output and code generation. We demonstrate that the current generation of LLMs performs poorly on these tasks and argue that there exists a significantly understudied domain in the development of code-based models, which involves incorporating the runtime context.

Accepted to the third Deep Learning for Code (DL4C) workshop @ ICLR 2025