Inner Speech as Behavior Guides: Steerable Imitation of Diverse Behaviors for Human-AI coordination

2026-02-24

AI Summary

Existing imitation learning approaches struggle to capture the diversity and non-Markovian nature of human behavior and lack flexible mechanisms for guiding behavior generation at inference time. This work proposes MIMIC, a novel framework that, for the first time, integrates human inner speech—a form of covert self-directed language—into AI behavior generation. By treating inner speech as a controllable linguistic representation of behavioral intent, MIMIC combines vision-language models, conditional variational autoencoders, and diffusion-based behavioral cloning to establish an inner-speech-mediated imitation learning architecture. The method enables fine-grained behavioral control during inference without requiring additional demonstrations, significantly improving behavioral diversity, fidelity to demonstrations, and controllability of behavioral style in robotic manipulation and human-robot collaboration tasks.

Technology Category

• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Effective human-AI coordination requires artificial agents capable of exhibiting and responding to human-like behaviors while adapting to changing contexts. Imitation learning has emerged as one of the prominent approaches to build such agents by training them to mimic human-demonstrated behaviors. However, current methods struggle to capture the inherent diversity and non-Markovian nature of human behavior and lack the ability to steer behavior at inference time. Drawing inspiration from the theory of human cognitive processes, where inner speech guides action selection before execution, we propose MIMIC (Modeling Inner Motivations for Imitation and Control), a framework that uses language as an internal representation of behavioral intent. MIMIC employs the novel use of vision-language models as linguistic scaffolding to train a conditional variational autoencoder capable of generating inner speech from observations. A diffusion-based behavior cloning policy then selects actions conditioned on current observations and the generated inner speech. MIMIC enables fine-grained steering of behavior at inference time by conditioning the agent on behavior-specific speech. Experiments across robotic manipulation tasks and human-AI collaboration games demonstrate that MIMIC significantly enhances both behavior diversity and fidelity to human demonstrations while enabling nuanced behavioral steering without training on additional demonstrations. We open source our code and provide pre-trained MIMIC agents and qualitative demos at: https://mimic-research.github.io.

Problem

• human-AI coordination
• imitation learning
• behavior diversity
• non-Markovian behavior
• behavior steering

Innovation

• inner speech
• imitation learning
• vision-language models
• conditional variational autoencoder
• diffusion policy