Infer Human's Intentions Before Following Natural Language Instructions

TL;DR

This paper introduces FISER, a framework that explicitly infers human goals and intentions through social reasoning to improve natural language instruction following in embodied tasks, outperforming existing methods.

Contribution

FISER models human internal goals as partially observable factors, enabling better reasoning and action planning in collaborative environments, achieving state-of-the-art results.

Findings

FISER outperforms end-to-end approaches in instruction following.

Explicit social reasoning improves understanding of human intentions.

Achieves state-of-the-art on the HandMeThat benchmark.

Abstract

For AI agents to be helpful to humans, they should be able to follow natural language instructions to complete everyday cooperative tasks in human environments. However, real human instructions inherently possess ambiguity, because the human speakers assume sufficient prior knowledge about their hidden goals and intentions. Standard language grounding and planning methods fail to address such ambiguities because they do not model human internal goals as additional partially observable factors in the environment. We propose a new framework, Follow Instructions with Social and Embodied Reasoning (FISER), aiming for better natural language instruction following in collaborative embodied tasks. Our framework makes explicit inferences about human goals and intentions as intermediate reasoning steps. We implement a set of Transformer-based models and evaluate them over a challenging benchmark, HandMeThat. We empirically demonstrate that using social reasoning to explicitly infer human intentions before making action plans surpasses purely end-to-end approaches. We also compare our implementation with strong baselines, including Chain of Thought prompting on the largest available pre-trained language models, and find that FISER provides better performance on the embodied social reasoning tasks under investigation, reaching the state-of-the-art on HandMeThat.