mlsanityData to 14 Oct 2026

ACES: Generating Diverse Programming Puzzles with with Autotelic Generative Models

arXiv 2310.10692 ↗First posted 15 Oct 2023cs.LG cs.AIcited by 6

Not on the map. Why?

Abstract (arXiv)

The ability to invent novel and interesting problems is a remarkable feature of human intelligence that drives innovation, art, and science. We propose a method that aims to automate this process by harnessing the power of state-of-the-art generative models to produce a diversity of challenging yet solvable problems, here in the context of Python programming puzzles. Inspired by the intrinsically motivated literature, Autotelic CodE Search (ACES) jointly optimizes for the diversity and difficulty of generated problems. We represent problems in a space of LLM-generated semantic descriptors describing the programming skills required to solve them (e.g. string manipulation, dynamic programming, etc.) and measure their difficulty empirically as a linearly decreasing function of the success rate of Llama-3-70B, a state-of-the-art LLM problem solver. ACES iteratively prompts a large language model to generate difficult problems achieving a diversity of target semantic descriptors (goal-directed exploration) using previously generated problems as in-context examples. ACES generates problems that are more diverse and more challenging than problems produced by baseline methods and three times more challenging than problems found in existing Python programming benchmarks on average across 11 state-of-the-art code LLMs.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 37 of its 58 references that are in the map. First posted on arXiv 15 Oct 2023.

Builds on

37 of 58 references.

References · 37

  1. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,320
  3. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  5. 2021Program Synthesis with Large Language Modelscited by 3,637
  6. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
  7. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,191
  8. 2023Code Llama: Open Foundation Models for Codecited by 2,139
  9. 2023Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generationcited by 1,391
  10. 2017Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithmcited by 734
  11. 2023WizardCoder: Empowering Code Large Language Models with Evol-Instructcited by 733
  12. 2023Textbooks Are All You Needcited by 569
25 more
  1. 2023TinyStories: How Small Can Language Models Be and Still Speak Coherent English?cited by 449
  2. 2015Illuminating search spaces by mapping elitescited by 425
  3. 2022Unnatural Instructions: Tuning Language Models with (Almost) No Human Laborcited by 397
  4. 2018Visual Reinforcement Learning with Imagined Goalscited by 251
  5. 2023Guiding Pretraining in Reinforcement Learning with Large Language Modelscited by 187
  6. 2020Automatic Curriculum Learning For Deep RL: A Short Surveycited by 150
  7. 2017Intrinsic Motivation and Automatic Curricula via Asymmetric Self-Playcited by 141
  8. 2022Evolution through Large Modelscited by 134
  9. 2023EvoPrompting: Language Models for Code-Level Neural Architecture Searchcited by 131
  10. 2023Language Model Crossover: Variation through Few-Shot Promptingcited by 122
  11. 2017Quality and Diversity Optimization: A Unifying Modular Frameworkcited by 113
  12. 2023Motif: Intrinsic Motivation from Artificial Intelligence Feedbackcited by 102
  13. 2023LLMatic: Neural Architecture Search via Large Language Models and Quality Diversity Optimizationcited by 87
  14. 2020Autotelic Agents with Intrinsically Motivated Goal-Conditioned Reinforcement Learning: a Short Surveycited by 85
  15. 2017Intrinsically Motivated Goal Exploration Processes with Automatic Curriculum Learningcited by 83
  16. 2022Language Models Can Teach Themselves to Program Bettercited by 78
  17. 2023OMNI: Open-endedness via Models of human Notions of Interestingnesscited by 70
  18. 2016Using Centroidal Voronoi Tessellations to Scale Up the Multi-dimensional Archive of Phenotypic Elites Algorithmcited by 66
  19. 2013Active Learning of Inverse Models with Intrinsically Motivated Goal Exploration in Robotscited by 58
  20. 2023Augmenting Autotelic Agents with Large Language Modelscited by 30
  21. 2021Programming Puzzlescited by 25
  22. 2018Curiosity Driven Exploration of Learned Disentangled Goal Spacescited by 16
  23. 2019Intrinsically Motivated Discovery of Diverse Patterns in Self-Organizing Systemscited by 13
  24. 2020Hierarchically Organized Latent Modules for Exploratory Search in Morphogenetic Systemscited by 13
  25. 2022Learning One Abstract Bit at a Time Through Self-Invented Experiments Encoded as Neural Networkscited by 4

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 5.

6 papers.

  1. 2024Surveying the Effects of Quality, Diversity, and Complexity in Synthetic Data From Large Language Modelscited by 35
  2. 2025Assessing Adaptive World Models in Machines with Novel Gamescited by 20
  3. 2025Automated Capability Discovery via Foundation Model Self-Explorationcited by 10
  4. 2025SPARQ: Synthetic Problem Generation for Reasoning via Quality-Diversity Algorithmscited by 7
  5. 2025ThrowBench: Benchmarking LLMs by Predicting Runtime Exceptionscited by 2
  6. 2025Optimizing for Persuasion Improves LLM Generalization: Evidence from Quality-Diversity Evolution of Debate Strategiescited by 0