Q-STRUM Debate: Query-Driven Contrastive Summarization for Recommendation Comparison
George-Kirollos Saad, Scott Sanner · 2025
Read Q-STRUM Debate: Query-Driven Contrastive Summarization for Recommendation Comparison by George-Kirollos Saad, Scott Sanner (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2502-12921.txt.
Description
George-Kirollos Saad and Scott Sanner stage debates between AI models to explain shopping choices in a 2025 paper. When a user asks for a product in plain words, the best picks are items the user has never seen, so the reasons for each pick are opaque. Careful contrast summaries can open that black box by stressing how items differ on the points the user named.
Past summarizers fell short of sharp, query-focused contrasts. The new method prompts two model sides to debate, which forces each contrast to be argued, not just listed. Across three data sets, the debate summaries beat prior methods by a clear margin. Shoppers gain reasons they can check, and sellers gain a fairer way to present trade-offs.
Catalog record
| CREATOR | George-Kirollos Saad, Scott Sanner |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2502-12921