A Comprehensive Comparison of Text Summarization Performance: Amultifaceted Evaluation of Larg Language Models With Practicalconsiderations
Year: 2026
Type: conference-paper
Abstract: Text summarization is crucial for mitigating information overload across domains. This research evaluates summarization performance across 17 large language models using seven diverse datasets at three output lengths (50, 100, 150 tokens). We employ a novel multi-dimensional framework assessing factual consistency, semantic similarity, lexical overlap, and human-like evaluation while considering both quality and efficiency factors. Key findings reveal significant differences between models, with specific models excelling in factual accuracy (deepseek-v3), human-like quality (claude-3-5-sonnet), processing efficiency (gemini-1.5-flash), and cost effectiveness (gemini-1.5-flash). Performance varies dramatically by dataset, with models struggling on technical domains but performing well on conversational content. We identified a critical tension between factual consistency (best at 50 tokens) and perceived quality (best at 150 tokens).
Authors: Anantharaman Janakiraman, Behnaz Ghoraani
Institution: Florida Atlantic University
Language: English
Topic: Topic Modeling
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
OpenAlex ID: W7130656132