Towards Stable and Personalised Profiles for Lexical Alignment in Spoken Human-Agent Dialogue

2025-09-04

AI Summary

This study addresses lexical alignment in spoken human–machine dialogue—enabling LLM-based conversational agents to dynamically adapt to users’ idiosyncratic lexical preferences for enhanced naturalness and communicative efficiency. To overcome instability in low-resource settings and redundancy in profile construction exhibited by existing approaches, we propose a compact, part-of-speech–aware personalized lexical profile. Leveraging transcribed spontaneous speech, our method allocates vocabulary items differentially by POS category (e.g., 5 adjectives, 10 adverbs), balancing data efficiency with representational stability. Experiments demonstrate that high-stability, high-coverage profiles can be constructed from just ten minutes of speech. Our approach significantly outperforms baselines across recall, coverage, and cosine similarity metrics. The resulting lightweight, interpretable, and scalable lexical alignment framework advances personalized dialogue systems under resource constraints.

Technology Category

• Natural Language Processing: Lexical Semantics and Morphology
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Lexical alignment, where speakers start to use similar words across conversation, is known to contribute to successful communication. However, its implementation in conversational agents remains underexplored, particularly considering the recent advancements in large language models (LLMs). As a first step towards enabling lexical alignment in human-agent dialogue, this study draws on strategies for personalising conversational agents and investigates the construction of stable, personalised lexical profiles as a basis for lexical alignment. Specifically, we varied the amounts of transcribed spoken data used for construction as well as the number of items included in the profiles per part-of-speech (POS) category and evaluated profile performance across time using recall, coverage, and cosine similarity metrics. It was shown that smaller and more compact profiles, created after 10 min of transcribed speech containing 5 items for adjectives, 5 items for conjunctions, and 10 items for adverbs, nouns, pronouns, and verbs each, offered the best balance in both performance and data efficiency. In conclusion, this study offers practical insights into constructing stable, personalised lexical profiles, taking into account minimal data requirements, serving as a foundational step toward lexical alignment strategies in conversational agents.

Problem

• Constructing stable personalized lexical profiles for agents
• Optimizing profile size and data efficiency for alignment
• Enabling lexical alignment in human-agent dialogue systems

Innovation

• Constructs stable personalized lexical profiles
• Uses minimal transcribed speech data
• Optimizes part-of-speech item quantities