Inner Speech as Behavior Guides: Steerable Imitation of Diverse Behaviors for Human-AI coordination
This addresses the challenge of improving human-AI coordination by allowing agents to mimic and adapt human behaviors more effectively, though it appears incremental as it builds on existing imitation learning methods with novel language integration.
The paper tackles the problem of enabling artificial agents to exhibit diverse and steerable human-like behaviors for human-AI coordination by proposing MIMIC, a framework that uses language as internal intent representation. The result shows significant enhancements in behavior diversity and fidelity to human demonstrations, with experiments in robotic manipulation and collaboration games demonstrating this without additional training.
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.