Simple Baselines are Competitive with Code Evolution
cs.AI, cs.LG
Submitted: 2026-02-18
Updated: 2026-10-12
Code: https://github.com/YonatanGideoni/code_evo_simple_baselines
Sources
- MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering
- Evaluating Large Language Models Trained on Code
- Inefficiencies of Meta Agents for Agent Design
- Mathematical exploration and discovery at scale
- How Should We Meta-Learn Reinforcement Learning Algorithms?
- Automated Design of Agentic Systems
- MLAgentBench: Evaluating Language Agents on Machine Learning Experimentation
- AIDE: AI-Driven Exploration in the Space of Code
- ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
- Eureka: Human-Level Reward Design via Coding Large Language Models
- Kosmos: An AI Scientist for Autonomous Discovery
- AlphaEvolve: A coding agent for scientific and algorithmic discovery
- Agent Laboratory: Using LLM Agents as Research Assistants
- AI Research Agents for Machine Learning: Search, Exploration, and Generalization in MLE-bench
- OpenHands: An Open Platform for AI Software Developers as Generalist Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems