AssistedDS: Benchmarking How External Domain Knowledge Assists LLMs in Automated Data Science

2025-05-25

arXiv.org

AI Summary

Large language models (LLMs) struggle to critically leverage external domain knowledge in automated data science. Method: We introduce AssistedDS—the first benchmark for domain-knowledge-assisted evaluation—comprising synthetic and real Kaggle datasets paired with beneficial or adversarial domain documents. Our “interpretable synthesis + real-world scenarios” dual-track framework employs multi-stage prompting to assess end-to-end capabilities: document retrieval, knowledge filtering, code generation, and execution validation. Contribution/Results: We uncover a critical “blind adoption” flaw in LLMs: they fail significantly in time-series modeling, cross-fold consistency, and categorical variable handling. Experiments show state-of-the-art models suffer sharp performance degradation under adversarial documents; beneficial knowledge fails to mitigate harmful information, exposing severe deficiencies in domain knowledge discrimination and robust application.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Application Domains: Humanities & Computational Social Science

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Large language models (LLMs) have advanced the automation of data science workflows. Yet it remains unclear whether they can critically leverage external domain knowledge as human data scientists do in practice. To answer this question, we introduce AssistedDS (Assisted Data Science), a benchmark designed to systematically evaluate how LLMs handle domain knowledge in tabular prediction tasks. AssistedDS features both synthetic datasets with explicitly known generative mechanisms and real-world Kaggle competitions, each accompanied by curated bundles of helpful and adversarial documents. These documents provide domain-specific insights into data cleaning, feature engineering, and model selection. We assess state-of-the-art LLMs on their ability to discern and apply beneficial versus harmful domain knowledge, evaluating submission validity, information recall, and predictive performance. Our results demonstrate three key findings: (1) LLMs frequently exhibit an uncritical adoption of provided information, significantly impairing their predictive performance when adversarial content is introduced, (2) helpful guidance is often insufficient to counteract the negative influence of adversarial information, and (3) in Kaggle datasets, LLMs often make errors in handling time-series data, applying consistent feature engineering across different folds, and interpreting categorical variables correctly. These findings highlight a substantial gap in current models' ability to critically evaluate and leverage expert knowledge, underscoring an essential research direction for developing more robust, knowledge-aware automated data science systems.

Problem

• Evaluating how LLMs leverage external domain knowledge in data science workflows
• Assessing LLMs' ability to discern helpful versus harmful domain knowledge
• Identifying gaps in LLMs' critical evaluation of expert knowledge

Innovation

• Benchmark evaluates LLMs using domain knowledge documents
• Tests include helpful and adversarial knowledge for tabular tasks
• Measures critical evaluation of external information in workflows