Conversational Alignment with Artificial Intelligence in Context

TL;DR

This paper introduces the CONTEXT-ALIGN framework to evaluate how well AI conversational agents adhere to human communication norms, highlighting current limitations of large language models in achieving full alignment.

Contribution

It proposes a new framework for assessing conversational alignment in AI and analyzes the limitations of current large language models in this context.

Findings

The CONTEXT-ALIGN framework formalizes human conversational norms.

Current LLM architectures face fundamental limitations in alignment.

The framework aids in evaluating AI design choices for better alignment.

Abstract

The development of sophisticated artificial intelligence (AI) conversational agents based on large language models raises important questions about the relationship between human norms, values, and practices and AI design and performance. This article explores what it means for AI agents to be conversationally aligned to human communicative norms and practices for handling context and common ground and proposes a new framework for evaluating developers' design choices. We begin by drawing on the philosophical and linguistic literature on conversational pragmatics to motivate a set of desiderata, which we call the CONTEXT-ALIGN framework, for conversational alignment with human communicative practices. We then suggest that current large language model (LLM) architectures, constraints, and affordances may impose fundamental limitations on achieving full conversational alignment.