AgenticDataBench: A Comprehensive Benchmark for Data Agents

2026-07-01

AI Summary

This work addresses the lack of fine-grained evaluation frameworks for large language model (LLM)-driven data agents across diverse data science tasks. To bridge this gap, the study introduces data science skills as a core evaluation dimension, leveraging task clustering from Stack Overflow to derive skill labels. It constructs a comprehensive benchmark spanning 15 vertical domains by integrating real-world B2B datasets with LLM-generated tasks. This benchmark enables systematic and granular assessment of data agents’ capabilities. Through an open-source evaluation platform, the authors conduct a thorough analysis of prominent data agents, uncovering significant performance disparities across distinct skill dimensions.

Abstract

Data science aims to derive actionable insights from heterogeneous raw data, unlocking the value of the massive amounts of data generated in modern society. Automating this process is essential to reducing labor-intensive efforts for data scientists and enabling scalable data-driven applications. Recently, large language model (LLM)-based data agents have emerged as a promising solution to automate data science workflows. However, the field lacks comprehensive benchmarks to rigorously evaluate these agents across diverse scenarios with fine-grained granularity. To address this gap, we propose AgenticDataBench, a comprehensive benchmark featuring realistic tasks spanning diverse domains with fine-grained ground-truth labels. This enables evaluations to capture the diversity and complexity of data science workflows and the detailed performance of agents. First, to cover diverse domains, we collect real datasets and tasks from 15 vertical domains, including 5 real-world B2B use cases from a leading fintech company. Second, to remove redundancy in real-world tasks and generate high-quality tasks for domains lacking real data, we introduce data science skills, recurring data-centric operational patterns, and quantify benchmark coverage by the number of skills included. Representative skills are extracted from large-scale task solutions on Stack Overflow using skill-aligned hierarchical clustering. Third, for real-world business tasks, we select task-solution pairs that maximize diversity in skill composition, ensuring broad coverage of practical scenarios. Fourth, to generate realistic tasks for devise domains without real tasks, we propose a systematic LLM-based task generation approach to create workflows and tasks based on these skills. Finally, we evaluate state-of-the-art data agents using our annotated benchmark and open-sourced testbed, providing detailed skill-level insights.

Problem

• data agents
• benchmark
• large language models
• data science workflows
• evaluation

Innovation

• data agents
• comprehensive benchmark
• data science skills
• LLM-based task generation
• skill-aligned clustering