DataSciBench: An LLM Agent Benchmark for Data Science

• URL: http://arxiv.org/abs/2502.13897v1
• Date: Wed, 19 Feb 2025 17:31:51 GMT
• Title: DataSciBench: An LLM Agent Benchmark for Data Science
• Authors: Dan Zhang, Sining Zhoubian, Min Cai, Fengzu Li, Lekang Yang, Wei Wang, Tianjiao Dong, Ziniu Hu, Jie Tang, Yisong Yue,
• Abstract summary: DataSciBench is a benchmark for evaluating Large Language Model (LLM) capabilities in data science. We develop a semi-automated pipeline for generating ground truth (GT) and validating evaluation metrics. We propose an innovative Task - Function - Code framework to assess each code execution outcome.
• Score: 33.3811507234528
• License:
• Abstract: This paper presents DataSciBench, a comprehensive benchmark for evaluating Large Language Model (LLM) capabilities in data science. Recent related benchmarks have primarily focused on single tasks, easily obtainable ground truth, and straightforward evaluation metrics, which limits the scope of tasks that can be evaluated. In contrast, DataSciBench is constructed based on a more comprehensive and curated collection of natural and challenging prompts for uncertain ground truth and evaluation metrics. We develop a semi-automated pipeline for generating ground truth (GT) and validating evaluation metrics. This pipeline utilizes and implements an LLM-based self-consistency and human verification strategy to produce accurate GT by leveraging collected prompts, predefined task types, and aggregate functions (metrics). Furthermore, we propose an innovative Task - Function - Code (TFC) framework to assess each code execution outcome based on precisely defined metrics and programmatic rules. Our experimental framework involves testing 6 API-based models, 8 open-source general models, and 9 open-source code generation models using the diverse set of prompts we have gathered. This approach aims to provide a more comprehensive and rigorous evaluation of LLMs in data science, revealing their strengths and weaknesses. Experimental results demonstrate that API-based models outperform open-sourced models on all metrics and Deepseek-Coder-33B-Instruct achieves the highest score among open-sourced models. We release all code and data at https://github.com/THUDM/DataSciBench.

Related papers

• A Systematic Approach for Assessing Large Language Models' Test Case Generation Capability [0.8287206589886879]
  We propose the Generated Benchmark from Control-Flow Structure and Variable Usage Composition (GBCV) approach to evaluate large language models (LLMs) By leveraging basic control-flow structures and variable usage, GBCV provides a flexible framework to create a spectrum of programs ranging from simple to complex. Our findings indicate that GPT-4o performs better on complex program structures, while all models effectively detect boundary values in simple conditions but face challenges with arithmetic computations.
  arXiv  Detail & Related papers  (2025-02-05T03:51:44Z)
• Can Models Help Us Create Better Models? Evaluating LLMs as Data Scientists [41.94295877935867]
  We present a benchmark for large language models designed to tackle one of the most knowledge-intensive tasks in data science. We demonstrate that the FeatEng of our proposal can cheaply and efficiently assess the broad capabilities of LLMs.
  arXiv  Detail & Related papers  (2024-10-30T17:59:01Z)
• Efficacy of Synthetic Data as a Benchmark [3.2968976262860408]
  We investigate the effectiveness of generating synthetic data through large language models (LLMs) Our experiments show that while synthetic data can effectively capture performance of various methods for simpler tasks, it falls short for more complex tasks like named entity recognition. We propose a new metric called the bias factor, which evaluates the biases introduced when the same LLM is used to both generate benchmarking data and to perform the tasks.
  arXiv  Detail & Related papers  (2024-09-18T13:20:23Z)
• SELF-GUIDE: Better Task-Specific Instruction Following via Self-Synthetic Finetuning [70.21358720599821]
  Large language models (LLMs) hold the promise of solving diverse tasks when provided with appropriate natural language prompts. We propose SELF-GUIDE, a multi-stage mechanism in which we synthesize task-specific input-output pairs from the student LLM. We report an absolute improvement of approximately 15% for classification tasks and 18% for generation tasks in the benchmark's metrics.
  arXiv  Detail & Related papers  (2024-07-16T04:41:58Z)
• DiscoveryBench: Towards Data-Driven Discovery with Large Language Models [50.36636396660163]
  We present DiscoveryBench, the first comprehensive benchmark that formalizes the multi-step process of data-driven discovery. Our benchmark contains 264 tasks collected across 6 diverse domains, such as sociology and engineering. Our benchmark, thus, illustrates the challenges in autonomous data-driven discovery and serves as a valuable resource for the community to make progress.
  arXiv  Detail & Related papers  (2024-07-01T18:58:22Z)
• DCA-Bench: A Benchmark for Dataset Curation Agents [9.60250892491588]
  We propose a dataset curation agent benchmark, DCA-Bench, to measure large language models' capability of detecting hidden dataset quality issues. Specifically, we collect diverse real-world dataset quality issues from eight open dataset platforms as a testbed. The proposed benchmark can also serve as a testbed for measuring the capability of LLMs in problem discovery rather than just problem-solving.
  arXiv  Detail & Related papers  (2024-06-11T14:02:23Z)
• Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models [52.98743860365194]
  We propose a new fine-tuning method called Self-Play fIne-tuNing (SPIN) At the heart of SPIN lies a self-play mechanism, where the LLM refines its capability by playing against instances of itself. This sheds light on the promise of self-play, enabling the achievement of human-level performance in LLMs without the need for expert opponents.
  arXiv  Detail & Related papers  (2024-01-02T18:53:13Z)
• On the Evaluation and Refinement of Vision-Language Instruction Tuning Datasets [71.54954966652286]
  We try to evaluate the Vision-Language Instruction-Tuning (VLIT) datasets. We build a new dataset, REVO-LION, by collecting samples with higher SQ from each dataset. Remarkably, even with only half of the complete data, the model trained on REVO-LION can achieve the performance comparable to simply adding all VLIT datasets up.
  arXiv  Detail & Related papers  (2023-10-10T13:01:38Z)
• MLLM-DataEngine: An Iterative Refinement Approach for MLLM [62.30753425449056]
  We propose a novel closed-loop system that bridges data generation, model training, and evaluation. Within each loop, the MLLM-DataEngine first analyze the weakness of the model based on the evaluation results. For targeting, we propose an Adaptive Bad-case Sampling module, which adjusts the ratio of different types of data. For quality, we resort to GPT-4 to generate high-quality data with each given data type.
  arXiv  Detail & Related papers  (2023-08-25T01:41:04Z)
• From Quantity to Quality: Boosting LLM Performance with Self-Guided Data Selection for Instruction Tuning [52.257422715393574]
  We introduce a self-guided methodology for Large Language Models (LLMs) to autonomously discern and select cherry samples from open-source datasets. Our key innovation, the Instruction-Following Difficulty (IFD) metric, emerges as a pivotal metric to identify discrepancies between a model's expected responses and its intrinsic generation capability.
  arXiv  Detail & Related papers  (2023-08-23T09:45:29Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.