Back to AI News
arxiv

ABSClaw: From Invented Worlds to Reality-Grounded Social Simulation

The paper introduces ABSClaw, a zero-code agent that compiles natural-language requests into executable social simulations grounded in survey, census, geospatial, and real-network data. It expresses state transitions through 26 atomic operations and supports lifecycle-scoped skills without modifying its core. The authors also propose ABSBench, a benchmark of 193 real-world social-simulation tasks, and report that ABSClaw achieves the best overall performance among evaluated social simulators. Case studies indicate that grounding personas, environments, and interaction networks improves agreement with observed outcomes, and one runtime skill executes 13 population-scale simulations involving 25 million agents.

AI Brief

It offers a concrete framework and benchmark for making LLM-based social simulations more empirically grounded and reproducible.

arXiv:2611.24517v1 Announce Type: new Abstract: LLM-based artificial societies exhibit rich behavior but often lack scientific grounding: personas and environments are assumed, interaction networks are synthetic, and simulators are generated as free-form code, making results difficult to tie to real populations and sensitive to the backbone model. We argue that simulated worlds should be sampled from observed data, while simulation logic should be implemented by composing operations from a shared, task-independent vocabulary. In this paper, we introduce ABSClaw, a zero-code agent that compiles natural-language requests into executable, reality-grounded social simulations. ABSClaw grounds who acts, where they act, and with whom they interact in survey and census data, geospatial sources, and real networks. It expresses state transitions through 26 atomic operations. To support an open and extensible ecosystem, ABSClaw further allows capabilities to be added as lifecycle-scoped skills without modifying its core. Furthermore, we propose ABSBench, a benchmark of 193 real-world social-simulation tasks. On ABSBench, ABSClaw achieves the best overall performance among the evaluated social simulators. Three real-world case studies show that grounding personas, environments, and interaction networks each improves agreement with observed outcomes. With one runtime skill, ABSClaw executes 13 population-scale simulations, collectively involving 25 million agents. Together, these results demonstrate ABSClaw's real-world fidelity, extensibility, and scalability.
Be the first to rate this article.

Comments & Questions (0)

Captcha
Please be respectful — let's keep the conversation friendly.

No comments yet

Be the first to comment!