BLADE: Benchmarking Language Model Agents for Data-Driven Science

TL;DR

BLADE is a benchmark designed to evaluate language model agents' ability to perform complex, open-ended scientific data analysis tasks, highlighting their strengths and limitations in supporting data-driven discovery.

Contribution

The paper introduces BLADE, a comprehensive benchmark with datasets and evaluation methods for assessing language model agents in scientific data analysis.

Findings

Language models often perform basic analyses.

Data-interacting agents show more diverse analytical approaches.

Evaluation methods effectively match agent outputs to expert ground truth.

Abstract

Data-driven scientific discovery requires the iterative integration of scientific domain knowledge, statistical expertise, and an understanding of data semantics to make nuanced analytical decisions, e.g., about which variables, transformations, and statistical models to consider. LM-based agents equipped with planning, memory, and code execution capabilities have the potential to support data-driven science. However, evaluating agents on such open-ended tasks is challenging due to multiple valid approaches, partially correct steps, and different ways to express the same decisions. To address these challenges, we present BLADE, a benchmark to automatically evaluate agents' multifaceted approaches to open-ended research questions. BLADE consists of 12 datasets and research questions drawn from existing scientific literature, with ground truth collected from independent analyses by expert data scientists and researchers. To automatically evaluate agent responses, we developed corresponding computational methods to match different representations of analyses to this ground truth. Though language models possess considerable world knowledge, our evaluation shows that they are often limited to basic analyses. However, agents capable of interacting with the underlying data demonstrate improved, but still non-optimal, diversity in their analytical decision making. Our work enables the evaluation of agents for data-driven science and provides researchers deeper insights into agents' analysis approaches.