MOSAIC: Modular Opinion Summarization using Aspect Identification and Clustering

• URL: http://arxiv.org/abs/2603.19277v1
• Date: Sun, 01 Mar 2026 15:32:34 GMT
• Title: MOSAIC: Modular Opinion Summarization using Aspect Identification and Clustering
• Abstract summary: We propose a scalable, modular framework designed for industrial deployment that decomposes summarization into interpretable components.<n>We validate the practical impact of our approach through online A/B tests on live product pages, showing that surfacing intermediate outputs improves customer experience.<n>We introduce opinion clustering as a system-level component and show that it significantly enhances faithfulness, particularly under the noisy and redundant conditions typical of user reviews.
• Score: 0.7519872646378835
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Reviews are central to how travelers evaluate products on online marketplaces, yet existing summarization research often emphasizes end-to-end quality while overlooking benchmark reliability and the practical utility of granular insights. To address this, we propose MOSAIC, a scalable, modular framework designed for industrial deployment that decomposes summarization into interpretable components, including theme discovery, structured opinion extraction, and grounded summary generation. We validate the practical impact of our approach through online A/B tests on live product pages, showing that surfacing intermediate outputs improves customer experience and delivers measurable value even prior to full summarization deployment. We further conduct extensive offline experiments to demonstrate that MOSAIC achieves superior aspect coverage and faithfulness compared to strong baselines for summarization. Crucially, we introduce opinion clustering as a system-level component and show that it significantly enhances faithfulness, particularly under the noisy and redundant conditions typical of user reviews. Finally, we identify reliability limitations in the standard SPACE dataset and release a new open-source tour experience dataset (TRECS) to enable more robust evaluation.

Related papers

• ReviewGrounder: Improving Review Substantiveness with Rubric-Guided, Tool-Integrated Agents [50.27474750319121]
  Review text according to paper-specific rubrics derived from official guidelines, the paper's content, and human-written reviews.<n>We introduce REVIEWBENCH, a benchmark evaluating review text according to paper-specific rubrics derived from official guidelines, the paper's content, and human-written reviews.<n>We propose REVIEWGROUNDER, a rubric-guided, tool-integrated multi-agent framework that decomposes reviewing into drafting and grounding stages.
  arXiv  Detail & Related papers  (2026-04-15T16:33:04Z)
• RAIR: A Rule-Aware Benchmark Uniting Challenging Long-Tail and Visual Salience Subset for E-commerce Relevance Assessment [26.73200625292998]
  We propose Rule-Aware benchmark with Image for Relevance assessment(RAIR), a Chinese dataset derived from real-world scenarios.<n>RAIR established a standardized framework for relevance assessment and provides a set of universal rules, which forms the foundation for standardized evaluation.<n>We conducted experiments on RAIR using 14 open and closed-source models.
  arXiv  Detail & Related papers  (2025-12-31T16:09:08Z)
• End-to-End Aspect-Guided Review Summarization at Scale [0.0]
  We present a scalable large language model (LLM)-based system that combines aspect-based sentiment analysis (ABSA) with guided summarization to generate concise and interpretable product review summaries.<n>Our approach first extracts and consolidates aspect-sentiment pairs from individual reviews, selects the most frequent aspects for each product, and samples representative reviews accordingly.
  arXiv  Detail & Related papers  (2025-09-30T11:24:07Z)
• OpinioRAG: Towards Generating User-Centric Opinion Highlights from Large-scale Online Reviews [12.338320566839483]
  We study the problem of opinion highlights generation from large volumes of user reviews.<n>Existing methods either fail to scale or produce generic, one-size-fits-all summaries that overlook personalized needs.<n>We introduce OpinioRAG, a scalable, training-free framework that combines RAG-based evidence retrieval with LLMs to efficiently produce tailored summaries.
  arXiv  Detail & Related papers  (2025-08-30T00:00:34Z)
• Can Large Language Models be Effective Online Opinion Miners? [9.91172891027387]
  We introduce Online Opinion Mining Benchmark (OOMB), a novel dataset and evaluation protocol designed to assess the ability of large language models to mine opinions effectively.<n>OOMB provides extensive (entity, feature, opinion) annotations and a comprehensive opinion-centric summary that highlights key opinion topics within each content.<n>We conduct a comprehensive analysis of which aspects remain challenging and where LLMs exhibit adaptability, to explore whether they can effectively serve as opinion miners in realistic online scenarios.
  arXiv  Detail & Related papers  (2025-05-21T16:09:44Z)
• Decomposed Opinion Summarization with Verified Aspect-Aware Modules [82.38097397662436]
  We propose a domain-agnostic modular approach guided by review aspects.<n>We conduct experiments across datasets representing scientific research, business, and product domains.
  arXiv  Detail & Related papers  (2025-01-27T09:29:55Z)
• AMRFact: Enhancing Summarization Factuality Evaluation with AMR-Driven Negative Samples Generation [57.8363998797433]
  We propose AMRFact, a framework that generates perturbed summaries using Abstract Meaning Representations (AMRs) Our approach parses factually consistent summaries into AMR graphs and injects controlled factual inconsistencies to create negative examples, allowing for coherent factually inconsistent summaries to be generated with high error-type coverage.
  arXiv  Detail & Related papers  (2023-11-16T02:56:29Z)
• Towards Personalized Review Summarization by Modeling Historical Reviews from Customer and Product Separately [59.61932899841944]
  Review summarization is a non-trivial task that aims to summarize the main idea of the product review in the E-commerce website. We propose the Heterogeneous Historical Review aware Review Summarization Model (HHRRS) We employ a multi-task framework that conducts the review sentiment classification and summarization jointly.
  arXiv  Detail & Related papers  (2023-01-27T12:32:55Z)
• Evaluating Interactive Summarization: an Expansion-Based Framework [97.0077722128397]
  We develop an end-to-end evaluation framework for interactive summarization. Our framework includes a procedure of collecting real user sessions and evaluation measures relying on standards. All of our solutions are intended to be released publicly as a benchmark.
  arXiv  Detail & Related papers  (2020-09-17T15:48:13Z)
• AliExpress Learning-To-Rank: Maximizing Online Model Performance without Going Online [60.887637616379926]
  This paper proposes an evaluator-generator framework for learning-to-rank. It consists of an evaluator that generalizes to evaluate recommendations involving the context, and a generator that maximizes the evaluator score by reinforcement learning. Our method achieves a significant improvement in terms of Conversion Rate (CR) over the industrial-level fine-tuned model in online A/B tests.
  arXiv  Detail & Related papers  (2020-03-25T10:27:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.