CivWorldBench: Evaluating Language World Models in Structured Multi-Agent Sandbox Worlds
cs.CL
Submitted: 2026-10-12
Updated: 2026-10-12
Code: https://github.com/Leon-Francis/CivWorldBench
Project page: https://leon-francis.github.io/CivWorldBench
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- WorldPrediction: A Benchmark for High-level World Modeling and Long-horizon Procedural Planning
- A Definition and Roadmap for World Models
- Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
- Approaching Human-Level Forecasting with Language Models
- Can Large Language Models Infer Causation from Correlation?
- ForecastBench: A Dynamic Benchmark of AI Forecasting Capabilities
- Bridging the Agent-World Gap: Text World Models for LLM-based Agents
- Evaluating LLMs on Real-World Forecasting Against Expert Forecasters
- BALROG: Benchmarking Agentic LLM and VLM Reasoning On Games
- Generative Agents: Interactive Simulacra of Human Behavior
- AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society
- Cooperate or Collapse: Emergence of Sustainable Cooperation in a Society of LLM Agents
- CivRealm: A Learning and Reasoning Odyssey in Civilization for Decision-Making Agents
- Debugging code world models
- Evaluating the World Model Implicit in a Generative Model
- SocietyBench: Forecasting Counterfactual Social-World Evolution
- CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists
- LLM-as-a-Prophet: Understanding Predictive Intelligence with Prophet Arena
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering