DataSciBench: An LLM Agent Benchmark for Data Science
DataSciBench: бенчмарк LLM-агентов для анализа данных
2026-01-01
SCID: 54.1/xr3m6j6r
Discuss with AI
DataSciBench benchmarkIntention-Function-Code frameworkLLM agentsdata science evaluationsemi-automated ground truth generation
Figures from the paper
Abstract (AI)
This paper presents DataSciBench, a comprehensive benchmark for evaluating Large Language Models (LLMs) in data science.Unlike existing benchmarks limited to single task, simple evaluation metrics, and readily available ground truth (GT), DataSciBench is built on curated, natural, and challenging prompts with complex evaluation criteria and uncertain GT.To bridge the gap, we develop a semi-automated GT generation pipeline, integrating LLM-based selfconsistency and human verification to ensure accuracy, predefined task types, and aggregate functions (metrics).Furthermore, we introduce an innovative Intention-Function-Code (IFC) framework, assessing code execution outcomes through metrics and programmatic rules.Evaluating 26 models (8 API-based, 8 open-source general, 9 code generation, and 1 agentic models), our approach offers rigorous insights into LLM strengths and weaknesses.Experimental results show API-based models outperform open-source counterparts across all metrics, with DeepAnalyze-8B leading among open-sourced models.We release all code and data at https://github. com/THUDM/DataSciBench/.
Key Findings
1
API-based models outperform open-source models across all metrics, while DeepAnalyze-8B is the strongest open-source model.
2
DataSciBench introduces a comprehensive benchmark for evaluating LLMs on natural, challenging data-science prompts with complex criteria and uncertain ground truth.
3
Evaluation covers 26 models across API-based, open-source general, code-generation, and agentic categories.
4
The Intention-Function-Code framework evaluates code-execution outcomes using task metrics and programmatic rules.
5
The benchmark uses a semi-automated ground-truth pipeline combining LLM self-consistency and human verification to improve evaluation accuracy.
Research Object
Large Language Models (LLMs) for data science
Research Subject
Their data-science task performance, including code-execution outcomes, evaluated with complex metrics and programmatic criteria
Publication Details
Publication Date
2026-01-01
Journal
Publisher
ISSN
Cited by
85
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest