OmniSapiens: A Foundation Model for Social Behavior Processing via Heterogeneity-Aware Relative Policy Optimization

TL;DR

OmniSapiens introduces HARPO, a novel RL method that effectively learns from heterogeneous social behavioral data, resulting in a foundation model with superior multitask performance and robust reasoning capabilities.

Contribution

The paper presents HARPO, a heterogeneity-aware RL algorithm, and develops OmniSapiens-7B 2.0, a social behavior foundation model that outperforms existing models across multiple tasks.

Findings

OmniSapiens-7B 2.0 achieves up to +16.85% performance gains.

It demonstrates strong performance on multitask and held-out settings.

HARPO outperforms recent RL methods in heterogeneous behavioral tasks.

Abstract

To develop socially intelligent AI, existing approaches typically model human behavioral dimensions (e.g., affective, cognitive, or social attributes) in isolation. Although useful, task-specific modeling often increases training costs and limits generalization across behavioral settings. Recent reasoning RL methods facilitate training a single unified model across multiple behavioral tasks, but do not explicitly address learning across different heterogeneous behavioral data. To address this gap, we introduce Heterogeneity-Aware Relative Policy Optimization (HARPO), an RL method that balances leaning across heterogeneous tasks and samples. This is achieved by modulating advantages to ensure that no single task or sample carries disproportionate influence during policy optimization. Using HARPO, we develop and release Omnisapiens-7B 2.0, a foundation model for social behavior processing. Relative to existing behavioral foundation models, Omnisapiens-7B 2.0 achieves the strongest performance across behavioral tasks, with gains of up to +16.85% and +9.37% on multitask and held-out settings respectively, while producing more explicit and robust reasoning traces. We also validate HARPO against recent RL methods, where it achieves the most consistently strong performance across behavioral tasks.