Large Language Models for Token-Efficient and Semantic-Preserving Opinion Summarization

2026-07-12

AI Summary

This study addresses the challenge of efficiently generating faithful and balanced summaries from large-scale, redundant, and opinion-imbalanced user-generated text. To this end, the authors propose a novel framework that integrates multidimensional categorization (by sentiment and topic), hierarchical sampling, and tailored large language model (LLM) prompting. This approach significantly reduces token consumption while enhancing semantic fidelity, viewpoint balance, and content coverage in the resulting summaries. Experimental evaluations on Amazon, Tripadvisor, and X/Twitter datasets demonstrate that the method consistently outperforms both traditional AI baselines and standard LLM-based summarization techniques, achieving superior performance at substantially lower computational cost.

Abstract

Opinionated text - spanning product reviews, hotel feedback, and social posts - captures rich signals about user experiences, preferences, and concerns. However, the scale, redundancy, and imbalance of such corpora make it challenging to analyze opinions effectively, particularly when the goal is to generate summaries that remain faithful to the diversity of viewpoints expressed. This paper presents a framework that preserves semantics in LLM-based opinion summarization while minimizing token usage. We combine multidimensional classification (e.g., sentiment, topics) with a family of stratified sampling strategies to select compact yet representative subsets of opinions before prompting the LLM. Tailored prompts then produce balanced summaries that surface the salient aspects expressed in the opinions (e.g., strengths and weaknesses of products/hotels). Experiments on Amazon product reviews, Tripadvisor hotel reviews, and X/Twitter posts demonstrate that our method significantly reduces token usage and computational cost while consistently outperforming traditional AI-based and standard LLM summarization baselines in terms of content coverage, balance, and semantic preservation.

Problem

• opinion summarization
• semantic preservation
• token efficiency
• viewpoint diversity
• redundancy and imbalance

Innovation

• token-efficient summarization
• semantic preservation
• stratified sampling
• opinion summarization
• large language models