REVIEW 23 cited by
MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
MLE-STAR: Machine Learning Engineering Agent via Search and Targeted Refinement
read the original abstract
Agents based on large language models (LLMs) for machine learning engineering (MLE) can automatically implement ML models via code generation. However, existing approaches to build such agents often rely heavily on inherent LLM knowledge and employ coarse exploration strategies that modify the entire code structure at once. This limits their ability to select effective task-specific models and perform deep exploration within specific components, such as experimenting extensively with feature engineering options. To overcome these, we propose MLE-STAR, a novel approach to build MLE agents. MLE-STAR first leverages external knowledge by using a search engine to retrieve effective models from the web, forming an initial solution, then iteratively refines it by exploring various strategies targeting specific ML components. This exploration is guided by ablation studies analyzing the impact of individual code blocks. Furthermore, we introduce a novel ensembling method using an effective strategy suggested by MLE-STAR. Our experimental results show that MLE-STAR achieves medals in 64% of the Kaggle competitions on the MLE-bench Lite, significantly outperforming the best alternative.
Forward citations
Cited by 23 Pith papers
-
CADENZA: Compiling Natural-Language Intent into Task-Specific Operator DAGs for Semantic Query Processing
CADENZA compiles each semantic operator instance into an intent-specific TxRA DAG and selects an executable plan under quality–latency–cost preferences via seed synthesis, structural rewrites, semantic alternatives, d...
-
LongDS-Bench: On the Failure of Long-Horizon Agentic Data Analysis
LongDS benchmark shows state-of-the-art agents achieve only 48.45% accuracy on long-horizon data analysis tasks, with performance dropping 47 points from early to late turns and state-maintenance errors causing most failures.
-
Synthetic Sandbox for Training Machine Learning Engineering Agents
SandMLE generates micro-scale synthetic MLE environments from seed tasks to enable 13x faster on-policy RL training, delivering 20-67% gains over SFT on MLE-bench-lite and better generalization to new scaffolds.
-
KompeteAI: Accelerated Autonomous Multi-Agent System for End-to-End Pipeline Generation for Machine Learning Problems
KompeteAI accelerates AutoML pipeline evaluation 6.9 times and beats prior systems by 3% on MLE-Bench through candidate merging, external RAG, and predictive early scoring.
-
One Reflection Is Not Enough: Self-Correcting Autonomous Research via Multi-Hypothesis Failure Attribution
SAGE with MHFA improves failure recovery in autonomous research agents, raising metrics-bearing outputs from 42% to 92% on a 12-topic benchmark versus single-reflection baselines.
-
CADENZA: Compiling Natural-Language Intent into Task-Specific Operator DAGs for Semantic Query Processing
CADENZA introduces TxRA algebra and logical/physical planners to compile intents into optimized task DAGs, reporting up to 0.49 quality, 165.7x latency, and 310.3x cost gains on SemBench versus prior SQPE optimizers.
-
CADENZA: Compiling Natural-Language Intent into Task-Specific Operator DAGs for Semantic Query Processing
CADENZA introduces TxRA and dual planners to compile semantic operator intents into optimized task DAGs, claiming large gains in quality, latency, and cost on SemBench.
-
DataMaster: Data-Centric Autonomous AI Research
DataMaster autonomously optimizes data via tree search and shared memory, raising medal rate 32.27% on MLE-Bench Lite and beating the base instruct model on GPQA.
-
DataMaster: Data-Centric Autonomous AI Research
DataMaster deploys an AI agent to autonomously engineer data via tree search over external sources, shared candidate pools, and memory of past outcomes, yielding 32% higher medal rates on MLE-Bench Lite and a small GP...
-
Pioneer Agent: Continual Improvement of Small Language Models in Production
Pioneer Agent automates the full lifecycle of adapting and continually improving small language models via diagnosis-driven data synthesis and regression-constrained retraining, delivering gains of 1.6-83.8 points on ...
-
Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search
Gome reaches 35.1% any-medal rate on MLE-Bench by mapping reasoning to gradient-based updates, outperforming tree search once models are sufficiently capable.
-
iML: Executable, Problem-Grounded, and Broadly Exploratory Code-Driven AutoML
iML's contract-based modular code generation with runtime verification reports 85% valid submissions and a 45% medal rate on MLE-BENCH, ahead of MLZero, MLE-STAR, and AutoML-Agent.
-
Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents
A dual-loop LLM-agent system that autonomously generates, validates, and deploys recommendation-model improvements at YouTube, reportedly outperforming human engineering baselines.
-
Self-Evolving Recommendation System: End-To-End Autonomous Model Optimization With LLM Agents
LLM agents acting as ML engineers autonomously generated optimizer, architecture, and reward changes that produced small live metric gains at YouTube when deployed through a dual offline/online loop.
-
Can We Predict Before Executing Machine Learning Agents?
LLMs primed with verified data reports predict agent solution quality at 61.5% accuracy, powering a Predict-then-Verify agent that converges 6x faster than execution-only baselines.
-
Matryoshka Agent: Unfolding Sub-Agents for Long-Horizon Machine Learning Engineering
Matryoshka Agent’s orchestrator–sub-agent hierarchy plus tree-ranked RL raises MLE-Dojo HumanRank, letting a 4B orchestrator approach o4-mini and giving a 30B coder up to 36.7% relative gain.
-
Towards Persistent Case-Based Memory for Autonomous Data Science: A CBR-Augmented R&D-Agent with a Locally Deployable Small Language Model
CBR integration into R&D-Agent with Gemma 4 31B yields directionally higher accuracy and lower variance than baseline on one of two Kaggle competitions.
-
GEAR: Genetic AutoResearch for Agentic Code Evolution
GEAR applies genetic algorithms to maintain and evolve multiple research states in autonomous code agents, outperforming single-path baselines by continuing to discover improvements over extended runs.
-
EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale
EvoMaster is a self-evolving agent framework that achieves state-of-the-art results on scientific benchmarks by enabling iterative hypothesis refinement and knowledge accumulation across domains.
-
AceGRPO: Adaptive Curriculum Enhanced Group Relative Policy Optimization for Autonomous Machine Learning Engineering
AceGRPO trains 30B-parameter LLM agents to achieve 100% valid submissions and competitive performance on MLE-Bench-Lite through evolving data buffers and adaptive task sampling.
-
TusoAI: Agentic Optimization for Scientific Methods
TusoAI is an LLM-based agent that builds and iteratively optimizes domain-specific computational methods for scientific data analysis, outperforming expert baselines on RNA-seq denoising and earth monitoring while rep...
-
MLEvolve: A Self-Evolving Framework for Automated Machine Learning Algorithm Discovery
MLEvolve is a self-evolving multi-agent LLM system with Progressive MCGS, Retrospective Memory, and adaptive coding modes that reports SOTA medal and submission rates on MLE-Bench under a 12-hour budget while outperfo...
-
EvoMaster: A Foundational Evolving Agent Framework for Agentic Science at Scale
EvoMaster is an open-source agent framework whose iterative self-evolution loop achieves the highest average score on 9 of 10 scientific agent benchmarks compared to OpenHands, OpenClaw, and Codex.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.