Agents, open-endedness, AGI — page 2

• Open-Endedness is Essential for Artificial Superhuman Intelligence
  arxiv:2406.04268 · 2024-06-06 · text available
• Ouroboros: A Self-Developing Frontier Coding Agent with Reviewed Core Evolution
  arxiv:2608.08311 · 2026-08-08 · text available
• Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions
  arxiv:1901.01753 · 2019-01-07 · text available
• Propose, Solve, Verify: Self-Play Through Formal Verification
  arxiv:2512.18160 · 2025-12-20 · text available
• Robust agents learn causal world models
  openreview:pOoKI3ouv1 · 2024-02-16 · text available
• Self-Improvements in Modern Agentic Systems: A Survey
  arxiv:2607.13104 · 2026-07-14 · text available
• Self-Programming AI: Code-Learning Agents for Autonomous Refactoring and Architectural Evolution
  doi:10.21203/rs.3.rs-6688473/v1 · 2025-05-20 · text available
• ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution
  url:sakana.ai/shinka-evolve · 2025-09-17 · metadata only
• SPADE: Self-Play in Adaptive Synthetic Executable Environments
  arxiv:2608.19197 · 2026-08-19 · text available
• Toward Training Superintelligent Software Agents through Self-Play SWE-RL
  arxiv:2512.18552 · 2025-12-21 · text available
• Vending-Bench: A Benchmark for Long-Term Coherence of Autonomous Agents
  arxiv:2502.15840 · 2025-02-21 · text available
• What Does It Take to Be a Good AI Research Agent? Studying the Role of Ideation Diversity
  arxiv:2511.15593 · 2025-11-19 · text available