mlsanityData to 7 Oct 2026

Can Language Models Serve as Text-Based World Simulators?

arXiv 2406.06485 ↗First posted 10 Jun 2024cs.CL cs.AIcited by 50

Not on the map. Why?

Cited by 25 in World models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Virtual environments play a key role in benchmarking advances in complex planning and decision-making tasks but are expensive and complicated to build by hand. Can current language models themselves serve as world simulators, correctly predicting how actions change different world states, thus bypassing the need for extensive manual coding? Our goal is to answer this question in the context of text-based simulators. Our approach is to build and use a new benchmark, called ByteSized32-State-Prediction, containing a dataset of text game state transitions and accompanying game tasks. We use this to directly quantify, for the first time, how well LLMs can serve as text-based world simulators. We test GPT-4 on this dataset and find that, despite its impressive performance, it is still an unreliable world simulator without further innovations. This work thus contributes both new insights into current LLM's capabilities and weaknesses, as well as a novel benchmark to track future progress as new models appear.

Key facts. Cited by 50 papers in the map (data through 7 Oct 2026). Builds on 17 of its 21 references that are in the map. First posted on arXiv 10 Jun 2024.

Builds on

17 of 21 references.

References · 17

  1. 2023GPT-4 Technical Reportcited by 18,132
  2. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,143
  3. 2023Reasoning with Language Model is Planning with World Modelcited by 846
  4. 2023LLM+P: Empowering Large Language Models with Optimal Planning Proficiencycited by 496
  5. 2023On the Planning Abilities of Large Language Models : A Critical Investigationcited by 308
  6. 2018TextWorld: A Learning Environment for Text-based Gamescited by 304
  7. 2019Interactive Fiction Games: A Colossal Adventurecited by 165
  8. 2019Learning to Speak and Act in a Fantasy Text Adventure Gamecited by 122
  9. 2023From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thoughtcited by 118
  10. 2024WorldCoder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environmentcited by 99
  11. 2023Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modellingcited by 88
  12. 2020Graph Constrained Reinforcement Learning for Natural Language Action Spacescited by 80
5 more
  1. 2020Learning Dynamic Belief Graphs to Generalize on Text-Based Gamescited by 49
  2. 2023ByteSized32: A Corpus and Challenge Task for Generating Task-Specific World Models Expressed as Text Gamescited by 22
  3. 2021A Systematic Survey of Text Worlds as Embodied Natural Language Environmentscited by 17
  4. 2023Towards Generating Functionally Correct Code Edits from Natural Language Issue Descriptionscited by 14
  5. 2019Generating Interactive Worlds with Textcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 9; 2025: 17; 2026: 24.

50 papers.

Sort:

By area:

  1. 2024Web Agents with World Models: Learning and Leveraging Environment Dynamics in Web Navigationcited by 116
  2. 2025Vid2World: Crafting Video Diffusion Models to Interactive World Modelscited by 67
  3. 2026Agent World Model: Infinity Synthetic Environments for Agentic Reinforcement Learningcited by 51
  4. 2024Knowledge Mechanisms in Large Language Models: A Survey and Perspectivecited by 50
  5. 2025From Word to World: Can Large Language Models be Implicit Text-based World Models?cited by 40
  6. 2025RLVR-World: Training World Models with Reinforcement Learningcited by 38
  7. 2025Planning with Reasoning using Vision Language World Modelcited by 36
  8. 2024CityBench: Evaluating the Capabilities of Large Language Models for Urban Taskscited by 26
  9. 2025Text2World: Benchmarking Large Language Models for Symbolic World Model Generationcited by 25
  10. 2026Qwen-AgentWorld: Language World Models for General Agentscited by 22
  11. 2024Making Large Language Models into World Models with Precondition and Effect Knowledgecited by 21
  12. 2025Large language models for artificial general intelligence (AGI): A survey of foundational principles and approachescited by 18
38 more
  1. 2026Visual Generation Unlocks Human-Like Reasoning through Multimodal World Modelscited by 18
  2. 2024Generative Emergent Communication: Large Language Model is a Collective World Modelcited by 15
  3. 2025Dyna-Think: Synergizing Reasoning, Acting, and World Model Simulation in AI Agentscited by 14
  4. 2025Evolution and The Knightian Blindspot of Machine Learningcited by 11
  5. 2025Revisiting the Othello World Model Hypothesiscited by 11
  6. 2026EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RLcited by 11
  7. 2024LLM-Based World Models Can Make Decisions Solely, But Rigorous Evaluations are Neededcited by 10
  8. 2025R-WoM: Retrieval-augmented World Model For Computer-use Agentscited by 10
  9. 2024Enhancing Agent Learning through World Dynamics Modelingcited by 9
  10. 2026Toward Scalable Verifiable Reward: Proxy State-Based Evaluation for Multi-turn Tool-Calling LLM Agentscited by 9
  11. 2024Towards Building Specialized Generalist AI with System 1 and System 2 Fusioncited by 8
  12. 2025G-Sim: Generative Simulations with Large Language Models and Gradient-Free Calibrationcited by 8
  13. 2025Evolution of Cooperation in LLM-Agent Societies: A Preliminary Study Using Different Punishment Strategiescited by 6
  14. 2025Trajectory World Models for Heterogeneous Environmentscited by 4
  15. 2026Beyond State Consistency: Behavior Consistency in Text-Based World Modelscited by 4
  16. 2026From Pixels to States: Rethinking Interactive World Models as Game Enginescited by 4
  17. 2026Egocentric Bias in Vision-Language Modelscited by 3
  18. 2026Bridging the Agent-World Gap: Text World Models for LLM-based Agentscited by 3
  19. 2025Identifying Non-Replicable Social Science Studies with Language Modelscited by 2
  20. 2026C-World: A Computer Use Agent Environment Creatorcited by 2
  21. 2026ARTIS: Agentic Risk-Aware Test-Time Scaling via Iterative Simulationcited by 2
  22. 2026Reward Prediction with Factorized World Statescited by 2
  23. 2026World-State Transformations for Neuro-symbolic Interactive Storytellingcited by 2
  24. 2024BeSimulator: A Large Language Model Powered Text-based Behavior Simulatorcited by 1
  25. 2025Foundation Models as World Models: A Foundational Study in Text-Based GridWorldscited by 1
  26. 2025Bridging the Gap Between Multimodal Foundation Models and World Modelscited by 1
  27. 2026When Stories Evolve: Benchmarking LLM Storytelling Across Agent Architectures in Open-Ended World Simulationscited by 1
  28. 2026EnvCraft: Synthesizing Executable Environments in Agentic RL for Claw-like Agentcited by 1
  29. 2025ByteSized32Refactored: Towards an Extensible Interactive Text Games Corpus for LLM World Modeling and Evaluationcited by 0
  30. 2026Textual Planning with Explicit Latent Transitionscited by 0
  31. 2026From Control to Foresight: Simulation as a New Paradigm for Human-Agent Collaborationcited by 0
  32. 2026Absurd World: A Simple Yet Powerful Method to Absurdify the Real-world for Probing LLM Reasoning Capabilitiescited by 0
  33. 2026Training the Orchestrator: A Supervised Approach to End-to-End PDDL Planning with LLM Agentscited by 0
  34. 2026Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstrationcited by 0
  35. 2026OraclePhys: A Systematic Framework for LLM Fine-Tuning on Structural Mechanicscited by 0
  36. 2026WMLLM: Self-Evolving Optimization Agents via Predict-Then-Act World Modelingcited by 0
  37. 2026World Agent: Can Language Models Keep a World Running?cited by 0
  38. 2026When Do Causal World Models Help Modular LLM Agentscited by 0