Self-Evolving Algorithm-Design Agents: Escaping In-Context Evolutionary Stagnation via Population-Curated Policy Optimization
cs.AI, cs.NE
Submitted: 2026-09-30
Updated: 2026-09-30
Sources
- Efficient RL Training for LLMs with Experience Replay
- A General Language Assistant as a Laboratory for Alignment
- MaxProof: Scaling Mathematical Proof with Generative-Verifier RL and Population-Level Test-Time Scaling
- A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Adam: A Method for Stochastic Optimization
- RePO: Replay-Enhanced Policy Optimization
- Scientific Algorithm Discovery by Augmenting AlphaEvolve with Deep Research
- Advancing Automated Algorithm Design via Evolutionary Stagewise Design with LLMs
- AlphaEvolve: A coding agent for scientific and algorithmic discovery
- Learning by Distilling Context
- Gemma 4 Technical Report
- Qwen3 Technical Report
- Learning to Discover at Test Time
- KernelBench-Verified: Do LLM-Generated Kernels Actually Beat PyTorch?
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