aiqbench analysis
This research introduces macro2mind, a method that trains language models to simulate individual behavioral reasoning by leveraging aggregate data from prediction markets rather than relying solely on pretraining or limited annotations. By explicitly decomposing social behavior into group interactions and belief updates, the approach achieves state-of-the-art forecasting accuracy on Polymarket and successfully transfers zero-shot to various user simulation benchmarks. The study demonstrates a novel pathway for scaling behavioral diversity in AI agents using macro-level financial signals as supervision.
Learning to Simulate Individuals from Macro Social Signals
arXiv:2610.07062v1 Announce Type: cross Abstract: Large language models are increasingly used to simulate how individuals respond to new situations, yet the behavioral reasoning behind these responses is either inherited from pretraining or learned from individual-level annotations, which offer limited behavioral diversity and little supervision of the reasoning itself. We propose to learn behavioral reasoning from prediction markets, whose price trajectories record how populations respond to real-world events at scale. We introduce macro2mind, which trains a language model with GRPO using market signals. A social behavioral decomposition makes behavioral reasoning an explicit step of forecasting: the model infers representative groups of market participants, predicts how each interprets the news and updates its beliefs, reasons about their interactions, and aggregates these responses into a price. A hindsight-regret curriculum with difficulty-aware sampling focuses training on transitions where hindsight-identified groups substantially improve the forecast while prioritizing examples that remain learnable for the current policy. The learned reasoning applies to user simulation without further training. On SWM-Bench, macro2mind achieves state-of-the-art directional accuracy and correlation on Polymarket. Trained on market data, it transfers zero-shot to four user-simulation benchmarks (Humanual, OvertonBench, PRISM, and CAD) and has competitive performance among zero-shot methods. Used as a data generator, macro2mind also raises a downstream simulator's accuracy on unseen users by 15.5 points, outperforming data generated by its backbone by 13.2 points.
Abstract:Large language models are increasingly used to simulate how individuals respond to new situations, yet the behavioral reasoning behind these responses is either inherited from pretraining or learned from individual-level annotations, which offer limited behavioral diversity and little supervision of the reasoning itself. We propose to learn behavioral reasoning from prediction markets, whose price trajectories record how populations respond to real-world events at scale. We introduce macro2mind, which trains a language model with GRPO using market signals. A social behavioral decomposition makes behavioral reasoning an explicit step of forecasting: the model infers representative groups of market participants, predicts how each interprets the news and updates its beliefs, reasons about their interactions, and aggregates these responses into a price. A hindsight-regret curriculum with difficulty-aware sampling focuses training on transitions where hindsight-identified groups substantially improve the forecast while prioritizing examples that remain learnable for the current policy. The learned reasoning applies to user simulation without further training. On SWM-Bench, macro2mind achieves state-of-the-art directional accuracy and correlation on Polymarket. Trained on market data, it transfers zero-shot to four user-simulation benchmarks (Humanual, OvertonBench, PRISM, and CAD) and has competitive performance among zero-shot methods. Used as a data generator, macro2mind also raises a downstream simulator's accuracy on unseen users by 15.5 points, outperforming data generated by its backbone by 13.2 points.
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Social and Information Networks (cs.SI)
arXiv:2610.07062 [cs.LG]
(or arXiv:2610.07062v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.07062 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 322 words
Related coverage
SocioVerse2: A Longitudinal Dynamic Social Simulation Framework under a Human-AI Co-evolutionary Paradigm
arXiv · Sep 22, 2026
Encoded Coordination on the Open Web
LessWrong · Sep 23, 2026
Game-Theoretic Disempowerment: Loss of Control to Agencyless AI
LessWrong · Sep 27, 2026
From the Fluency Fallacy to the Micro-to-Macro Validity Gap: Opportunities and Pitfalls of LLMs in Social Simulation
arXiv · Sep 9, 2026