# SocietyBench: Forecasting Counterfactual Social-World Evolution

arXiv: https://arxiv.org/abs/2608.04009 · HTML: https://mlsanity.com/paper/2608.04009/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 23 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Builds on (most cited first, 21 of 21)

- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [GAIA: a benchmark for General AI Assistants](https://mlsanity.com/paper/2311.12983/) (2023)
- [AgentSociety: Large-Scale Simulation of LLM-Driven Generative Agents Advances Understanding of Human Behaviors and Society](https://mlsanity.com/paper/2502.08691/) (2025)
- [Simulating Opinion Dynamics with Networks of LLM-based Agents](https://mlsanity.com/paper/2311.09618/) (2023)
- [OASIS: Open Agent Social Interaction Simulations with One Million Agents](https://mlsanity.com/paper/2411.11581/) (2024)
- [ForecastBench: A Dynamic Benchmark of AI Forecasting Capabilities](https://mlsanity.com/paper/2409.19839/) (2024)
- [Approaching Human-Level Forecasting with Language Models](https://mlsanity.com/paper/2402.18563/) (2024)
- [Forecasting Future World Events with Neural Networks](https://mlsanity.com/paper/2206.15474/) (2022)
- [FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction](https://mlsanity.com/paper/2508.11987/) (2025)
- [SocioVerse: A World Model for Social Simulation Powered by LLM Agents and A Pool of 10 Million Real-World Users](https://mlsanity.com/paper/2504.10157/) (2025)
- [MIRAI: Evaluating LLM Agents for Event Forecasting](https://mlsanity.com/paper/2407.01231/) (2024)
- [MF-LLM: Simulating Population Decision Dynamics via a Mean-Field Large Language Model Framework](https://mlsanity.com/paper/2504.21582/) (2025)
- [PolyBench: Benchmarking LLM Forecasting and Trading Capabilities on Live Prediction Market Data](https://mlsanity.com/paper/2604.14199/) (2026)
- [Social Opinions Prediction Utilizes Fusing Dynamics Equation with LLM-based Agents](https://mlsanity.com/paper/2409.08717/) (2024)
- [DualMind: Towards Understanding Cognitive-Affective Cascades in Public Opinion Dissemination via Multi-Agent Simulation](https://mlsanity.com/paper/2602.02534/) (2026)
- [Context-Aware Sentiment Forecasting via LLM-based Multi-Perspective Role-Playing Agents](https://mlsanity.com/paper/2505.24331/) (2025)

## Built on by (most cited first, 1 of 1)

- [CivWorldBench: Evaluating Language World Models in Structured Multi-Agent Sandbox Worlds](https://mlsanity.com/paper/2610.16450/) (2026)

## Abstract (arXiv)

Large language models (LLMs), and the agents built on top of them, are now benchmarked heavily on whether they can finish a task -- fix a bug, drive a browser, operate a GUI. A complementary social ability, namely how well a model understands and forecasts the way real social events unfold, has barely been measured. We introduce SocietyBench, an end-to-end benchmark that takes a one-line event topic, collects Web news and social-media posts across five platforms, distills them into a date-indexed timeline that keeps factual events and a public-opinion layer separate, and then turns every cutoff date on that timeline into an audited bank of forecasting questions. Questions are scored on two orthogonal 100-point axes: probability calibration and temporal accuracy. Before any model sees a timeline, a three-phase procedure replaces every named entity and shifts every date by a per-event constant, turning a real arc into a counterfactual social world -- structurally identical to what happened, but stripped of the surface labels a model could match against pre-training memory. On five heterogeneous events and 125 prediction points in Chinese and English editions, the strongest of six frontier LLMs reaches only 75.0 out of 100, against a trivial anchor of 50. The two axes come apart: a model can be calibration-strong but time-weak, or the reverse. Three agent frameworks built on a shared base model fail to improve on that base, and two model-free heuristics trail every LLM. Per-event gaps reach 21.4 points on a single axis, which is our main argument for evaluating on several events rather than one. All anonymized timelines, question banks, ground truth, and scoring code are released.