Q-STRUM Debate: Query-Driven Contrastive Summarization for Recommendation Comparison

TL;DR

Q-STRUM Debate enhances query-driven contrastive summarization for recommendation comparison by employing debate-style prompting with large language models, significantly improving contrastive summaries over existing methods.

Contribution

It introduces a novel debate prompting approach for contrastive summarization, leveraging LLMs to generate more focused and effective contrastive summaries in recommendation systems.

Findings

Significant performance improvements over existing contrastive summarization methods.

Effective use of debate-style prompting with LLMs for better contrastive summaries.

Validated across three datasets with improved criteria.

Abstract

Query-driven recommendation with unknown items poses a challenge for users to understand why certain items are appropriate for their needs. Query-driven Contrastive Summarization (QCS) is a methodology designed to address this issue by leveraging language-based item descriptions to clarify contrasts between them. However, existing state-of-the-art contrastive summarization methods such as STRUM-LLM fall short of this goal. To overcome these limitations, we introduce Q-STRUM Debate, a novel extension of STRUM-LLM that employs debate-style prompting to generate focused and contrastive summarizations of item aspects relevant to a query. Leveraging modern large language models (LLMs) as powerful tools for generating debates, Q-STRUM Debate provides enhanced contrastive summaries. Experiments across three datasets demonstrate that Q-STRUM Debate yields significant performance improvements over existing methods on key contrastive summarization criteria, thus introducing a novel and performant debate prompting methodology for QCS.