Discovered, Not Designed: Population Evolution for Collaborative and Compute-Intensive Model Discovery
cs.LG
Submitted: 2026-10-05
Updated: 2026-10-05
Code: https://github.com/karpathy/autoresearch
Sources
- Optuna: A Next-generation Hyperparameter Optimization Framework
- PIQA: Reasoning about Physical Commonsense in Natural Language
- AdaEvolve: Adaptive LLM Driven Zeroth-Order Optimization
- MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- KuaiRand: An Unbiased Sequential Recommendation Dataset with Randomly Exposed Videos
- ResearchGym: Evaluating Language Model Agents on Real-World AI Research
- Gemma 3 Technical Report
- Evolutionary Multi-Task Optimization for LLM-Guided Program Discovery
- Qwen2.5-Coder Technical Report
- Averaging Weights Leads to Wider Optima and Better Generalization
- Population Based Training of Neural Networks
- LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
- ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
- MO-EMT-NAS: Multi-Objective Continuous Transfer of Architectural Knowledge Between Tasks from Different Datasets
- EvoX: Meta-Evolution for Automated Discovery
- Janus: An Algorithm-Evaluator Co-Evolution Framework for LLM-Driven Discovery under Expensive Evaluation Budgets
- The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery
- MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks