mlsanityData to 14 Oct 2026

Finite Automata Extraction: Low-data World Model Learning as Programs from Gameplay Video

arXiv 2508.11836 ↗First posted 15 Aug 2025cs.AIcited by 0

On the map: World models → Video-generation world models and simulators

Abstract (arXiv)

World models are defined as a compressed spatial and temporal learned representation of an environment. The learned representation is typically a neural network, making transfer of the learned environment dynamics and explainability a challenge. In this paper, we propose an approach, Finite Automata Extraction (FAE), that learns a neuro-symbolic world model from gameplay video represented as programs in a novel domain-specific language (DSL): Retro Coder. Compared to prior world model approaches, FAE learns a more precise model of the environment and more general code than prior DSL-based approaches.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Video-generation world models and simulators, one of 11 directions in World models. Builds on 14 of its 24 references that are in the map. First posted on arXiv 15 Aug 2025.

Builds on

14 of 24 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Diffusion Models Are Real-Time Game Engines27 Aug

Other references · 13

  1. 2018World Modelscited by 1,513
  2. 2020DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learningcited by 161
  3. 2016DeepCoder: Learning to Write Programscited by 149
  4. 2024iVideoGPT: Interactive VideoGPTs are Scalable World Modelscited by 146
  5. 2017RobustFill: Neural Program Learning under Noisy I/Ocited by 126
  6. 2019Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Surveycited by 93
  7. 2020Learning to Simulate Dynamic Environments with GameGANcited by 86
  8. 2021Playable Video Generationcited by 64
  9. 2021MarioNette: Self-Supervised Sprite Learningcited by 35
  10. 2019Few-Shot Bayesian Imitation Learning with Logical Program Policiescited by 30
  11. 2023Learning of Generalizable and Interpretable Knowledge in Grid-Based Reinforcement Learning Environmentscited by 5
  12. 2017Mechanics Automatically Recognized via Interactive Observation: Jumpingcited by 4
1 more
  1. 2024KnowPC: Knowledge-Driven Programmatic Reinforcement Learning for Zero-shot Coordinationcited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Code World Model: Coding Agent as World Brain
  2. 2026AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research
  3. 2024Pandora: Towards General World Model with Natural Language Actions and Video States
  4. 2026World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis
  5. 2025A Constructive Framework for Nondeterministic Automata via Time-Shared, Depth-Unrolled Feedforward Networks
  6. 2024Autoverse: An Evolvable Game Language for Learning Robust Embodied Agents
  7. 2026Code Plans, Diffusion Renders: Open-Ended Generative World Modeling
  8. 2024WorldDreamer: Towards General World Models for Video Generation via Predicting Masked Tokens