Literature

All papers, repositories, and systems covered in this survey. 41 papers, 52 repositories, 61 systems.

P02: General-Purpose Evolutionary Frameworks

SystemPaperRepositoryCreatorDate
LLM4ADarXiv:2412.17287Optima-CityU/LLM4ADCity University of Hong Kong2024-12
AlphaEvolvearXiv:2506.13131google-deepmind/alphaevolve_resultsGoogle DeepMind2025-05
OpenEvolve--algorithmicsuperintelligence/openevolveAlgorithmic SuperIntelligence Labs2025-05
GEPA: Optimize AnythingarXiv:2507.19457gepa-ai/gepaUC Berkeley / Stanford2025-07
ShinkaEvolvearXiv:2509.19349SakanaAI/ShinkaEvolveSakana AI2025-09
SkyDiscover & AdaEvolvearXiv:2602.20133skydiscover-ai/skydiscoverUC Berkeley Sky Computing Lab2026-02

P03: Self-Improving Agent Systems

SystemPaperRepositoryCreatorDate
Darwin Gödel MachinearXiv:2505.22954jennyzzt/dgmSakana AI2025-05
Darwinian Evolver--imbue-ai/darwinian_evolverImbue2026-02
GEPA SkillsarXiv:2507.19457gepa-ai/gepaUC Berkeley / Stanford2026-02
EvoSkillarXiv:2603.02766sentient-agi/EvoSkillSentient AGI2026-03
NeoSigma: Self-Improving Agentic Systemslink--NeoSigma2026-03
RetroAgent: Self-Evolving via Retrospective FeedbackarXiv:2603.08561zhangxy-2019/RetroAgentXiaoying Zhang et al.2026-03
AutoEvolver--tengxiaoliu/autoevolver----

P04: Program Synthesis & ARC-AGI Solvers

SystemPaperRepositoryCreatorDate
AB-MCTS & TreeQuestarXiv:2503.04412SakanaAI/treequestSakana AI2025-03
Confluence Labs--confluence-labs/arc-agi-2Confluence Labs2025-12
Arcgentica--symbolica-ai/arcgenticaSymbolica AI2025-12

P05: Benchmarks, Discovery & Applications

SystemPaperRepositoryCreatorDate
The AI ScientistarXiv:2408.06292SakanaAI/AI-ScientistSakana AI2024-08
ALE-BencharXiv:2506.09050SakanaAI/ALE-BenchSakana AI2025-06
ShinkaEvolve at ICFP 2025arXiv:2509.19349SakanaAI/ShinkaEvolveSakana AI2025-10
AlphaEvolve Research PapersarXiv:2511.02864--Google DeepMind2025-11
Matlantis Crystal Structure PredictionarXiv:2503.21201--Preferred Networks2026-01
AI Scientist: Nature PublicationarXiv:2408.06292SakanaAI/AI-ScientistSakana AI2026-03
DiscoGenarXiv:2603.17863AlexGoldie/discogenGoogle DeepMind2026-03
Meta-Harness: End-to-End Harness OptimizationarXiv:2603.28052stanford-iris-lab/meta-harness-tbench2-artifactStanford IRIS Lab2026-03

P06: Evolutionary Scaling & Efficiency

SystemPaperRepositoryCreatorDate
Evolution Strategies at ScalearXiv:1703.03864VsonicV/es-fine-tuning-paperOpenAI2017-03
EGGROLL: Hyperscale Evolution StrategiesarXiv:2511.16652ESHyperscale/HyperscaleESNVIDIA / University of Oxford2025-11
A-EvolvearXiv:2602.00359A-EVO-Lab/a-evolveA-EVO Lab2026-02
EvoX: Meta-EvolutionarXiv:2602.23413skydiscover-ai/skydiscoverUC Berkeley / Stanford2026-02

P07: Autonomous Research Systems

SystemPaperRepositoryCreatorDate
CycleResearcherarXiv:2411.00816zhu-minjun/ResearcherZhu et al.2024-11
AgentLaboratory: End-to-End Research WorkflowarXiv:2501.04227SamuelSchmidgall/AgentLaboratorySamuel Schmidgall et al.2025-01
AI Scientist v2arXiv:2504.08066SakanaAI/AI-Scientist-v2Sakana AI2025-04
AI-ResearcherarXiv:2505.18705HKUDS/AI-ResearcherHKUDS (University of Hong Kong)2025-05
Zochi--IntologyAI/ZochiIntologyAI2025-05
RD-Agent: Microsoft R&D Automation PlatformarXiv:2505.14738microsoft/RD-AgentMicrosoft2025-05
Karpathy Autoresearch--karpathy/autoresearchAndrej Karpathy2025-06
SkyPilot Autoresearch--skypilot-org/skypilotUC Berkeley2025-06
K-Dense Co-ScientistarXiv:2508.07043K-Dense-AI/k-dense-byokK-Dense AI2025-08
DeepScientistarXiv:2509.26603ResearAI/DeepScientistWestlake University2025-09
DeepResearch: Alibaba Autonomous Research AgentarXiv:2510.24701Alibaba-NLP/DeepResearchAlibaba (Qwen Team)2025-10
freephdlabor: Personalized 24/7 Research AgentarXiv:2510.15624ltjed/freephdlaborTsinghua University / Sakana AI2025-10
OmniScientist: Tsinghua AI Research EcosystemarXiv:2511.16931tsinghua-fib-lab/OmniScientistTsinghua University2025-11
EurekaClaw--EurekaClaw/EurekaClawEurekaClaw2026-01
AIRA₂arXiv:2603.26499--Meta2026-03
AutoResearchClaw--aiming-lab/AutoResearchClawAIMING Lab (UNC Chapel Hill)2026-03
Bilevel AutoresearcharXiv:2603.23420EdwardOptimization/Bilevel-AutoresearchIndependent Researchers2026-03
OpenResearcherarXiv:2603.20278TIGER-AI-Lab/OpenResearcherTIGER-AI-Lab (U of Waterloo)2026-03
Pi-Autoresearch--davebcn87/pi-autoresearchShopify2026-03
ASI-Evolve: AI Accelerates AIarXiv:2603.29640--Shanghai AI Lab2026-03
Sakana Marlin: Autonomous Research Agentlink--Sakana AI2026-03
NanoResearch: GPU-Native Autonomous Research--OpenRaiser/NanoResearchOpenRaiser2026-03
Omni-SimpleMem: Autoresearch-Guided Memory DiscoveryarXiv:2604.01007aiming-lab/SimpleMemAIMING Lab (UNC Chapel Hill)2026-04

P08: Harness & Agent Frameworks

SystemPaperRepositoryCreatorDate
Hyperspace--hyperspaceai/agiHyperspace AI2025-06
UlamAI--ulamai/ulamaiUlamAI2026-01
AutoHarnessarXiv:2603.03329--Google DeepMind2026-02
HyperagentsarXiv:2603.19461facebookresearch/HyperAgentsMeta2026-03
OpenProver--Kripner/openproverOpen Source2026-03
Ouro Loop--VictorVVedtion/ouro-loopOpen Source2026-03
AutoAgent: Hands-Off Agent Optimization--kevinrgu/autoagentKevin Gu2026-03
7/24 Office--wangziqi06/724-office----