Improving Language Model Prompting in Support of Semi-autonomous Task Learning

TL;DR

This paper explores methods for constructing prompts for language models to generate specific, actionable responses that aid agents in learning new tasks within their environment.

Contribution

It introduces a novel approach for prompt construction that produces context-specific, interpretable responses to support semi-autonomous task learning.

Findings

Actionable task knowledge can be effectively extracted from LLMs.

Prompting strategies significantly influence response relevance and usefulness.

Empirical results validate the approach for online agent learning.

Abstract

Language models (LLMs) offer potential as a source of knowledge for agents that need to acquire new task competencies within a performance environment. We describe efforts toward a novel agent capability that can construct cues (or "prompts") that result in useful LLM responses for an agent learning a new task. Importantly, responses must not only be "reasonable" (a measure used commonly in research on knowledge extraction from LLMs) but also specific to the agent's task context and in a form that the agent can interpret given its native language capacities. We summarize a series of empirical investigations of prompting strategies and evaluate responses against the goals of targeted and actionable responses for task learning. Our results demonstrate that actionable task knowledge can be obtained from LLMs in support of online agent task learning.