Exodia Distilling SotA open-endedness through open-endedness.
Last updated: 2026-10-10 · upstream jennyzzt/awesome-open-ended@966c9ec · run #29

State-of-the-art concepts

Derived from 103 curated works via tfidf_kmeans analysis.

Key concepts (curated, ranked by prevalence)

How many of the 103 curated works engage each open-endedness concept, detected across full text + abstracts.

Theme clusters (the "majority vote")

world, llm, grand

17 works · world, llm, grand, innovation, large language

17 papers in this cluster
  1. Word2World: Generating Stories and Worlds through Large Language Models
  2. Eureka: Human-Level Reward Design via Coding Large Language Models
  3. Practical PCG Through Large Language Models
  4. Voyager: An Open-Ended Embodied Agent with Large Language Models
  5. Dreaming in Code for Curriculum Learning in Open-Ended Worlds
  6. MineDojo: Building Open-Ended Embodied Agents with Internet-Scale Knowledge
  7. Motif: Intrinsic Motivation from Artificial Intelligence Feedback
  8. OS-Copilot: Towards Generalist Computer Agents with Self-Improvement
  9. Evolution through Large Models
  10. Debating with More Persuasive LLMs Leads to More Truthful Answers
  11. Watts: Infrastructure for Open-Ended Learning
  12. Open-Endedness, World Models, and the Automation of Innovation
  13. Open-endedness: The last grand challenge you’ve never heard of
  14. Novel Opportunities in Open-Endedness
  15. EvoCraft: A New Challenge for Open-Endedness
  16. The Future of AI is Open-Ended
  17. AI That Can Improve Itself

poet, solve, novelty

16 works · poet, solve, novelty, environment, game

16 papers in this cluster
  1. Co-generation of game levels and game-playing agents
  2. Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions
  3. Transfer Dynamics in Emergent Evolutionary Curricula
  4. Enhanced POET: Open-Ended Reinforcement Learning through Unbounded Invention of Learning Challenges and their Solutions
  5. Endlessly Generating Increasingly Complex and Diverse Learning Environments and their Solutions through the Paired Open-Ended Trailblazer (POET)
  6. Co-optimising Robot Morphology and Controller in a Simulated Open-Ended Environment
  7. Open Questions in Creating Safe Open-ended AI: Tensions Between Control and Creativity
  8. Open-ended search for environments and adapted agents using map-elites
  9. RAPid-Learn: A Framework for Learning to Recover for Handling Novelties in Open-World Environments
  10. OMNI-EPIC: Open-endedness via Models of human Notions of Interestingness with Environments Programmed in Code
  11. Language and Culture Internalisation for Human-Like Autotelic AI
  12. Artificial Generational Intelligence: Cultural Accumulation in Reinforcement Learning
  13. DreamCraft: Text-Guided Generation of Functional 3D Environments in Minecraft
  14. Exploiting Open-Endedness to Solve Problems Through the Search for Novelty
  15. Identifying Necessary Conditions for Open-Ended Evolution through the Artificial Life World of Chromaria
  16. Video Games as a Testbed for Open-Ended Phenomena

rl, environment design, agent

15 works · rl, environment design, agent, policy, regret

15 papers in this cluster
  1. Emergent Complexity and Zero-shot Transfer via Unsupervised Environment Design
  2. Replay-Guided Adversarial Environment Design
  3. Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design
  4. Grounding Aleatoric Uncertainty in Unsupervised Environment Design
  5. MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learning
  6. Evolving Curricula with Regret-Based Environment Design
  7. Reward-Free Curricula for Training Robust World Models
  8. MiniHack the Planet: A Sandbox for Open-Ended Reinforcement Learning Research
  9. Environment Generation for Zero-Shot Compositional Reinforcement Learning
  10. Open-Ended Learning Leads to Generally Capable Agents
  11. Human-Timescale Adaptation in an Open-Ended Task Space
  12. Prioritized Level Replay
  13. Evolution and The Knightian Blindspot of Machine Learning
  14. Executive Function: A Contrastive Value Policy for Resampling and Relabeling Perceptions via Hindsight Summarization?
  15. Flow-Lenia: Towards open-ended evolution in cellular automata through mass conservation and parameter localization

agent, foundation, era

15 works · agent, foundation, era, meta, machine

15 papers in this cluster
  1. Automated design of agentic systems
  2. Darwin Gödel Machine: Open-Ended Evolution of Self-Improving Agents
  3. The AI Scientist: Towards fully automated open-ended scientific discovery
  4. Hyperagents
  5. The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search
  6. Learning to Continually Learn via Meta-learning Agentic Memory Designs
  7. Gödel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement
  8. Automated Capability Discovery via Model Self-Exploration
  9. Intelligent Go-Explore: Standing on the Shoulders of Giant Foundation Models
  10. AI-GAs: AI-generating algorithms, an alternate paradigm for producing general artificial intelligence
  11. Augmentative Topology Agents For Open-Ended Learning
  12. Powderworld: A Platform for Understanding Generalization via Rich Task Distributions
  13. OMNI: Open-endedness via Models of human Notions of Interestingness
  14. Open-ended and AI-generating algorithms in the era of foundation models
  15. Computational Life: How Well-formed, Self-replicating Programs Emerge from Simple Interaction

intelligence, objective, discovery

13 works · intelligence, objective, discovery, representation, research

13 papers in this cluster
  1. Beyond Fixed Representations: The Vocabulary and Verifier Gaps in Open-Ended AI
  2. Safety is Essential for Responsible Open-Ended Systems
  3. General Intelligence Requires Rethinking Exploration
  4. A Compositional Framework for Open-ended Intelligence
  5. Open-Endedness is Essential for Artificial Superhuman Intelligence
  6. Managing extreme AI risks amid rapid progress
  7. Questioning Representational Optimism in Deep Learning: The Fractured Entangled Representation Hypothesis
  8. Creative Problem Solving in Artificially Intelligent Agents: A Survey and Framework
  9. CORAL: Towards Autonomous Multi-Agent Evolution for Open-Ended Discovery
  10. Bootstrapping Intelligence: Self-Improving Systems for Open-Ended Discovery
  11. DREAM Architecture: a Developmental Approach to Open-Ended Learning in Robotics
  12. Structurally Flexible Neural Networks: Evolving the Building Blocks for General Agents
  13. Why Greatness Cannot Be Planned: The Myth of the Objective

llm, prompt, state art

11 works · llm, prompt, state art, teaming, identifying

11 papers in this cluster
  1. Rainbow Teaming: Open-Ended Generation of Diverse Adversarial Prompts
  2. Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
  3. Evolutionary Optimization of Model Merging Recipes
  4. Discovering Preference Optimization Algorithms with and for Large Language Models
  5. AgenticRed: Optimizing Agentic Systems for Automated Red-teaming
  6. DéjàQ: Open-Ended Evolution of Diverse, Learnable and Verifiable Problems
  7. Multi-Agent Diagnostics for Robustness via Illuminated Diversity
  8. Deep Laplacian-based Options for Temporally-Extended Exploration
  9. SPOTTER: Extending Symbolic Planning Operators through Targeted Reinforcement Learning
  10. Evolving Prompts In-Context: An Open-ended, Self-replicating Perspective
  11. Identifying Life-Changing Books with LLMs

quality diversity, qd, solution

9 works · quality diversity, qd, solution, search, language

9 papers in this cluster
  1. Quality-Diversity through AI Feedback
  2. Quality Diversity through Human Feedback
  3. Large Language Models as In-context AI Generators for Quality-Diversity
  4. Toward Artificial Open-Ended Evolution within Lenia using Quality-Diversity
  5. Improving Deep Reinforcement Learning via Quality Diversity, Open-Ended and AI-Generating Algorithms
  6. Foundation Model Self-Play: Open-Ended Strategy Innovation via Foundation Models
  7. In Search of the Ingredients of Open-Endedness: Replicating Picbreeder with Large Vision-Language Models
  8. Augmenting Autotelic Agents with Large Language Models
  9. Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation

Dominant vocabulary (consensus keyphrases)

  1. agent (2.1954)
  2. environment (1.8121)
  3. llm (1.3945)
  4. world (1.3573)
  5. search (1.3125)
  6. language (1.3118)
  7. design (1.2638)
  8. reinforcement (1.2541)
  9. diverse (1.2405)
  10. exploration (1.1975)
  11. large (1.1359)
  12. domain (1.1216)
  13. foundation (1.1091)
  14. diversity (1.0779)
  15. training (1.0573)
  16. game (1.0437)
  17. generate (1.0123)
  18. intelligence (1.0106)
  19. evocraft (1.0)
  20. discovery (0.9818)