Active Intelligence in Video Avatars via Closed-loop World Modeling

TL;DR

This paper introduces ORCA, a novel framework for active, goal-directed video avatars that use a closed-loop world model to adaptively interact with their environment, enabling autonomous multi-step task completion.

Contribution

We propose ORCA, the first framework integrating internal world modeling with a hierarchical dual-system architecture for active, goal-oriented video avatars in stochastic environments.

Findings

ORCA outperforms open-loop baselines in task success rate.

It achieves higher behavioral coherence in complex scenarios.

The framework demonstrates effective continuous belief updating and outcome verification.

Abstract

Current video avatar generation methods excel at identity preservation and motion alignment but lack genuine agency, they cannot autonomously pursue long-term goals through adaptive environmental interaction. We address this by introducing L-IVA (Long-horizon Interactive Visual Avatar), a task and benchmark for evaluating goal-directed planning in stochastic generative environments, and ORCA (Online Reasoning and Cognitive Architecture), the first framework enabling active intelligence in video avatars. ORCA embodies Internal World Model (IWM) capabilities through two key innovations: (1) a closed-loop OTAR cycle (Observe-Think-Act-Reflect) that maintains robust state tracking under generative uncertainty by continuously verifying predicted outcomes against actual generations, and (2) a hierarchical dual-system architecture where System 2 performs strategic reasoning with state prediction while System 1 translates abstract plans into precise, model-specific action captions. By formulating avatar control as a POMDP and implementing continuous belief updating with outcome verification, ORCA enables autonomous multi-step task completion in open-domain scenarios. Extensive experiments demonstrate that ORCA significantly outperforms open-loop and non-reflective baselines in task success rate and behavioral coherence, validating our IWM-inspired design for advancing video avatar intelligence from passive animation to active, goal-oriented behavior.