mlsanityData to 14 Oct 2026

MLZero: A Multi-Agent System for End-to-end Machine Learning Automation

arXiv 2505.13941 ↗First posted 20 May 2025cs.MA cs.AI cs.CLcited by 24

On the map: LLM agents and tool use → Web, search, and research agents

Abstract (arXiv)

Existing AutoML systems have advanced the automation of machine learning (ML); however, they still require substantial manual configuration and expert input, particularly when handling multimodal data. We introduce MLZero, a novel multi-agent framework powered by Large Language Models (LLMs) that enables end-to-end ML automation across diverse data modalities with minimal human intervention. A cognitive perception module is first employed, transforming raw multimodal inputs into perceptual context that effectively guides the subsequent workflow. To address key limitations of LLMs, such as hallucinated code generation and outdated API knowledge, we enhance the iterative code generation process with semantic and episodic memory. MLZero demonstrates superior performance on MLE-Bench Lite, outperforming all competitors in both success rate and solution quality, securing six gold medals. Additionally, when evaluated on our Multimodal AutoML Agent Benchmark, which includes 25 more challenging tasks spanning diverse data modalities, MLZero outperforms the competing methods by a large margin with a success rate of 0.92 (+263.6%) and an average rank of 2.28. Our approach maintains its robust effectiveness even with a compact 8B LLM, outperforming full-size systems from existing solutions.

Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use. Builds on 37 of its 84 references that are in the map. First posted on arXiv 20 May 2025.

Builds on

37 of 84 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering9 Oct
  2. 2024Data Interpreter: An LLM Agent For Data Science28 Feb
  3. 2024DS-Agent: Automated Data Science by Empowering Large Language Models with Case-Based Reasoning27 Feb
  4. 2023MLAgentBench: Evaluating Language Agents on Machine Learning Experimentation5 Oct

Other references · 33

  1. 2023GPT-4 Technical Reportcited by 18,396
  2. 2024The Llama 3 Herd of Modelscited by 14,715
  3. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  4. 2021Program Synthesis with Large Language Modelscited by 3,637
  5. 2023ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIscited by 1,947
  6. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,375
  7. 2023StarCoder: may the source be with you!cited by 799
  8. 2024Chronos: Learning the Language of Time Seriescited by 759
  9. 2024A Survey on Large Language Models for Code Generationcited by 658
  10. 2022MTEB: Massive Text Embedding Benchmarkcited by 640
  11. 2023MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Actioncited by 517
  12. 2020AutoGluon-Tabular: Robust and Accurate AutoML for Structured Datacited by 383
21 more
  1. 2022Large Language Models are Better Reasoners with Self-Verificationcited by 307
  2. 2025AIDE: AI-Driven Exploration in the Space of Codecited by 226
  3. 2020CLIMATE-FEVER: A Dataset for Verification of Real-World Climate Claimscited by 196
  4. 2023Planning with Large Language Models for Code Generationcited by 182
  5. 2024AutoML-Agent: A Multi-Agent LLM Framework for Full-Pipeline AutoMLcited by 87
  6. 2025Lifelong Learning of Large Language Model based Agents: A Roadmapcited by 83
  7. 2023Using Large Language Models for Hyperparameter Optimizationcited by 71
  8. 2024AutoKaggle: A Multi-Agent Framework for Autonomous Data Science Competitionscited by 70
  9. 2024Convolution Meets LoRA: Parameter Efficient Finetuning for Segment Anything Modelcited by 69
  10. 2025OctoTools: An Agentic Framework with Extensible Tools for Complex Reasoningcited by 69
  11. 2023Tool Documentation Enables Zero-Shot Tool-Usage with Large Language Modelscited by 68
  12. 2024Large Language Model Agent for Hyper-Parameter Optimizationcited by 64
  13. 2023Large Language Models as Data Preprocessorscited by 34
  14. 2023TabRepo: A Large Scale Repository of Tabular Model Evaluations and its AutoML Applicationscited by 33
  15. 2024AutoGluon-Multimodal (AutoMM): Supercharging Multimodal AutoML with Foundation Modelscited by 30
  16. 2023LLM Performance Predictors are good initializers for Architecture Searchcited by 29
  17. 2024Kolb-Based Experiential Learning for Generalist Agents with Human-Level Kaggle Data Science Performancecited by 25
  18. 2024SELA: Tree-Search Enhanced LLM Agents for Automated Machine Learningcited by 24
  19. 2025I-MCTS: Enhancing Agentic AutoML via Introspective Monte Carlo Tree Searchcited by 14
  20. 2018Statistical Analysis on E-Commerce Reviews, with Sentiment Classification using Bidirectional Recurrent Neural Network (RNN)cited by 8
  21. 2025AutoML Benchmark with shorter time constraints and early stoppingcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 19.

24 papers.

Sort:

By area:

  1. 2025Vibe Coding vs. Agentic Coding: Fundamentals and Practical Implications of Agentic AIcited by 42
  2. 2026MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discoverycited by 28
  3. 2025AutoMLGen: Navigating Fine-Grained Optimization for Coding Agentscited by 18
  4. 2026Frontis-MA1: Training an AI4AI Model towards Recursive Self-Improvement in Machine Learning Engineeringcited by 8
  5. 2026Learning to Ideate for Machine Learning Engineering Agentscited by 7
  6. 2026DataMaster: Data-Centric Autonomous AI Researchcited by 5
  7. 2026Why Solve It Twice? Hierarchical Accumulation of Skills for Transfer-Efficient ML Engineeringcited by 3
  8. 2026CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experimentscited by 2
  9. 2026Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineeringcited by 2
  10. 2026Exploit More, Explore Smarter for Budget-Constrained Agentic Searchcited by 2
  11. 2025LightAutoDS-Tab: Multi-AutoML Agentic System for Tabular Datacited by 1
  12. 2025Talking Trees: Reasoning-Assisted Induction of Decision Trees for Tabular Datacited by 1
12 more
  1. 2025SOCRATES: Simulation Optimization with Correlated Replicas and Adaptive Trajectory Evaluationscited by 1
  2. 2026stratum: A System Infrastructure for Massive Agent-Centric ML Workloadscited by 1
  3. 2026AgentGA: Evolving Code Solutions in Agent-Seed Spacecited by 1
  4. 2026CANTANTE: Optimizing Agentic Systems via Contrastive Credit Attributioncited by 1
  5. 2026Be Fair! Can Machine Learning Engineering Agents Adhere to Fairness Constraints?cited by 1
  6. 2026iML: Executable, Problem-Grounded, and Broadly Exploratory Code-Driven AutoMLcited by 0
  7. 2026A Framework for Assessing AI Agent Decisions and Outcomes in AutoML Pipelinescited by 0
  8. 2026Token Signatures of Code: Comparing Coding Behaviors Across Large Language Modelscited by 0
  9. 2026Strategy Accumulation and Guided Execution for Automated LLM Fine-Tuningcited by 0
  10. 2026Prompt2Skill: Unsupervised Skill Optimization From Natural Language Instructionscited by 0
  11. 2026How Much of a Harness Does a Strong Agent Need for Autonomous ML Engineering?cited by 0
  12. 2026ML Engineering Agents - A Data Management Perspectivecited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024BudgetMLAgent: A Cost-Effective LLM Multi-Agent system for Automating Machine Learning Tasks
  2. 2025MLE-Smith: Scaling MLE Tasks with Automated Multi-Agent Pipeline
  3. 2025ML-Agent: Reinforcing LLM Agents for Autonomous Machine Learning Engineering
  4. 2025MLE-Dojo: Interactive Environments for Empowering LLM Agents in Machine Learning Engineering
  5. 2024DuetML: Human-LLM Collaborative Machine Learning Framework for Non-Expert Users
  6. 2025NNGPT: Rethinking AutoML with Large Language Models
  7. 2025AutoMind: Adaptive Knowledgeable Agent for Automated Data Science
  8. 2024MLLM-Tool: A Multimodal Large Language Model For Tool Agent Learning