LLM-Based Data Science Agents: A Survey of Capabilities, Challenges, and Future Directions

• URL: http://arxiv.org/abs/2510.04023v1
• Date: Sun, 05 Oct 2025 04:04:27 GMT
• Title: LLM-Based Data Science Agents: A Survey of Capabilities, Challenges, and Future Directions
• Abstract summary: This survey presents the first comprehensive, lifecycle-aligned taxonomy of data science agents.<n>We analyze forty-five systems onto the six stages of the end-to-end data science process.<n>We highlight strengths and limitations at each stage, and review emerging benchmarks and evaluation practices.
• Score: 46.70253280146778
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Recent advances in large language models (LLMs) have enabled a new class of AI agents that automate multiple stages of the data science workflow by integrating planning, tool use, and multimodal reasoning across text, code, tables, and visuals. This survey presents the first comprehensive, lifecycle-aligned taxonomy of data science agents, systematically analyzing and mapping forty-five systems onto the six stages of the end-to-end data science process: business understanding and data acquisition, exploratory analysis and visualization, feature engineering, model building and selection, interpretation and explanation, and deployment and monitoring. In addition to lifecycle coverage, we annotate each agent along five cross-cutting design dimensions: reasoning and planning style, modality integration, tool orchestration depth, learning and alignment methods, and trust, safety, and governance mechanisms. Beyond classification, we provide a critical synthesis of agent capabilities, highlight strengths and limitations at each stage, and review emerging benchmarks and evaluation practices. Our analysis identifies three key trends: most systems emphasize exploratory analysis, visualization, and modeling while neglecting business understanding, deployment, and monitoring; multimodal reasoning and tool orchestration remain unresolved challenges; and over 90% lack explicit trust and safety mechanisms. We conclude by outlining open challenges in alignment stability, explainability, governance, and robust evaluation frameworks, and propose future research directions to guide the development of robust, trustworthy, low-latency, transparent, and broadly accessible data science agents.

Related papers

• Agents in the Wild: Where Research Meets Deployment [11.36937211823131]
  Agentic systems large language model (LLM) based architectures capable of reasoning, planning, acting, and coordinating with tools are rapidly transitioning from research prototypes to production scale deployments across domains such as software engineering, scientific discovery, and finance.<n>This tutorial brings together researchers and practitioners to explore advances in reasoning and planning, multi agent coordination, and evaluation, highlighting open challenges arising from deployment experience.
  arXiv  Detail & Related papers  (2026-07-21T17:55:10Z)
• Towards Robotic Dexterous Hand Intelligence: A Survey [64.42459413339479]
  Robotic dexterous hands are central to contact-rich manipulation, with rapid progress driven by advances in hardware, sensing, control, simulation, and data generation.<n>Existing studies are often developed under different assumptions regarding hand embodiments, sensory configurations, task settings, training data, and evaluation protocols.<n>This survey provides a holistic review of dexterous hand research from four complementary aspects.
  arXiv  Detail & Related papers  (2026-05-13T15:23:17Z)
• The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration [58.61149924275458]
  Tool use enables large language models to access external information, invoke software systems, and act in digital environments beyond what can be solved from model parameters alone.<n>As agent systems evolve, the central problem has shifted from isolated invocation to multi-tool orchestration over long trajectories with intermediate state, execution feedback, changing environments, and practical constraints such as safety, cost, and verifiability.<n>We organize the literature around six core dimensions: inference-time planning and execution, training and trajectory construction, safety and control, efficiency under resource constraints, capability completeness in open environments, and benchmark design and evaluation.
  arXiv  Detail & Related papers  (2026-03-24T07:05:05Z)
• Agentic Reasoning for Large Language Models [122.81018455095999]
  Reasoning is a fundamental cognitive process underlying inference, problem-solving, and decision-making.<n>Large language models (LLMs) demonstrate strong reasoning capabilities in closed-world settings, but struggle in open-ended and dynamic environments.<n>Agentic reasoning marks a paradigm shift by reframing LLMs as autonomous agents that plan, act, and learn through continual interaction.
  arXiv  Detail & Related papers  (2026-01-18T18:58:23Z)
• AI Agent Systems: Architectures, Applications, and Evaluation [4.967019713320407]
  AI agents combine foundation models with reasoning, planning, memory, and tool use.<n>We organize prior work into a unified taxonomy spanning agent components.<n>We discuss key design trade-offs -- latency vs. accuracy, autonomy vs. controllability, and capability vs. reliability.
  arXiv  Detail & Related papers  (2026-01-05T02:38:40Z)
• FinSight: Towards Real-World Financial Deep Research [68.31086471310773]
  FinSight is a novel framework for producing high-quality, multimodal financial reports.<n>To ensure professional-grade visualization, we propose an Iterative Vision-Enhanced Mechanism.<n>A two-stage Writing Framework expands concise Chain-of-Analysis segments into coherent, citation-aware, and multimodal reports.
  arXiv  Detail & Related papers  (2025-10-19T14:05:35Z)
• LLM/Agent-as-Data-Analyst: A Survey [51.19078438787228]
  Large language model (LLM) and agent techniques for data analysis have demonstrated substantial impact in both academica and industry.<n>The technical evolution further distills five key design goals for intelligent data analysis agents, namely semantic-aware design, hybrid integration, autonomous pipelines, tool-augmented modality, and support for open-world tasks.
  arXiv  Detail & Related papers  (2025-09-28T17:31:38Z)
• AI Agentic Programming: A Survey of Techniques, Challenges, and Opportunities [8.086360127362815]
  Large language model (LLM)-based coding agents autonomously plan, execute, and interact with tools such as compilers, debuggers, and version control systems.<n>Unlike conventional code generation, these agents decompose goals, coordinate multi-step processes, and adapt based on feedback, reshaping software development practices.
  arXiv  Detail & Related papers  (2025-08-15T00:14:31Z)
• A Survey of AI for Materials Science: Foundation Models, LLM Agents, Datasets, and Tools [15.928285656168422]
  Foundation models (FMs) are enabling scalable, general-purpose, and multimodal AI systems for scientific discovery.<n>This survey provides a comprehensive overview of foundation models, agentic systems, datasets, and computational tools supporting this growing field.
  arXiv  Detail & Related papers  (2025-06-25T18:10:30Z)
• Large Language Models Meet Stance Detection: A Survey of Tasks, Methods, Applications, Challenges and Future Directions [0.37865171120254354]
  Stance detection is essential for understanding subjective content across various platforms such as social media, news articles, and online reviews.<n>Recent advances in Large Language Models (LLMs) have revolutionized stance detection by introducing novel capabilities.<n>We present a novel taxonomy for LLM-based stance detection approaches, structured along three key dimensions.<n>Key applications in stance detection, political analysis, public health monitoring, and social media moderation are discussed.
  arXiv  Detail & Related papers  (2025-05-13T11:47:49Z)
• Survey on Evaluation of LLM-based Agents [28.91672694491855]
  The emergence of LLM-based agents represents a paradigm shift in AI.<n>This paper provides the first comprehensive survey of evaluation methodologies for these increasingly capable agents.
  arXiv  Detail & Related papers  (2025-03-20T17:59:23Z)
• Deep Learning, Machine Learning, Advancing Big Data Analytics and Management [26.911181864764117]
  Advances in artificial intelligence, machine learning, and deep learning have catalyzed the transformation of big data analytics and management.<n>This work explores the theoretical foundations, methodological advancements, and practical implementations of these technologies.<n>It equips researchers, practitioners, and data enthusiasts with the tools to navigate the complexities of modern data analytics.
  arXiv  Detail & Related papers  (2024-12-03T05:59:34Z)
• Data Analysis in the Era of Generative AI [56.44807642944589]
  This paper explores the potential of AI-powered tools to reshape data analysis, focusing on design considerations and challenges. We explore how the emergence of large language and multimodal models offers new opportunities to enhance various stages of data analysis workflow. We then examine human-centered design principles that facilitate intuitive interactions, build user trust, and streamline the AI-assisted analysis workflow across multiple apps.
  arXiv  Detail & Related papers  (2024-09-27T06:31:03Z)
• The Responsible Foundation Model Development Cheatsheet: A Review of Tools & Resources [100.23208165760114]
  Foundation model development attracts a rapidly expanding body of contributors, scientists, and applications.<n>To help shape responsible development practices, we introduce the Foundation Model Development Cheatsheet.
  arXiv  Detail & Related papers  (2024-06-24T15:55:49Z)

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.