We Argue to Agree: Towards Personality-Driven Argumentation-Based Negotiation Dialogue Systems for Tourism
This work addresses the need for more adaptive and personalized negotiation systems in tourism, though it is incremental as it builds on existing argumentation and personality integration methods.
The authors tackled the problem of enhancing negotiation dialogue systems by integrating personality-driven argumentation, proposing a new task and dataset for tourism negotiations. They demonstrated that fine-tuned LLMs effectively generate personalized rational responses, with evaluations showing high-quality dialogues and improved personalization.
Integrating argumentation mechanisms into negotiation dialogue systems improves conflict resolution through exchanges of arguments and critiques. Moreover, incorporating personality attributes enhances adaptability by aligning interactions with individuals' preferences and styles. To advance these capabilities in negotiation dialogue systems, we propose a novel Personality-driven Argumentation-based Negotiation Dialogue Generation (PAN-DG) task. To support this task, we introduce PACT, a dataset of Personality-driven Argumentation-based negotiation Conversations for Tourism sector. This dataset, generated using Large Language Models (LLMs), features three distinct personality profiles, viz. Argumentation Profile, Preference Profile, and Buying Style Profile to simulate a variety of negotiation scenarios involving diverse personalities. Thorough automatic and manual evaluations indicate that the dataset comprises high-quality dialogues. Further, we conduct comparative experiments between pre-trained and fine-tuned LLMs for the PAN-DG task. Multi-dimensional evaluation demonstrates that the fine-tuned LLMs effectively generate personality-driven rational responses during negotiations. This underscores the effectiveness of PACT in enhancing personalization and reasoning capabilities in negotiation dialogue systems, thereby establishing a foundation for future research in this domain.