[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85608-en":3,"doc-seo-85608-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85608,3848291630094,"Emma Wilson","https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45",8,"Research & Report","A Multi-Model Metric-based Selection Framework for Abstractive Text Summarization","Rapid digital text growth makes automatic abstractive summarization increasingly important for turning long articles into concise, informative outputs. This paper proposes a Multi-Model Summarization Framework that improves robustness and summary quality by generating candidate summaries with multiple fine-tuned transformer models and then selecting the best one via metric-based evaluation. Candidates are scored for lexical similarity and semantic relevance using automatic metrics, and the highest-quality summary is chosen. Fine-tuning and evaluation on CNN/DailyMail show the top BERTScore (88.63%) and strong performance versus several LLM baselines.","A Multi-Model Metric-based Selection Framework for Abstractive Text summarization  \nAhmed Alansary  \nFaculty of Computer Science MSA University Giza, Egypt  \n[ahmed.mohamed406@msa.edu.eg](ahmed.mohamed406@msa.edu.eg)  \nAli Hamdi  \nFaculty of Computer Science MSA University Giza, Egypt [ahamdi@msa.edu.eg](ahamdi@msa.edu.eg)  \narXiv :2606 .05494v 3 [ cs .CL] 13 Jul 2026  \nAbstract—Automatic text summarization has become increasingly important due to the rapid growth of digital textual information. This paper presents a Multi-Model Summarization Framework designed to improve the robustness and quality of abstractive text summarization. Relying on a single model often leads to inconsistent summarization quality across articles with varying structures and topics. To address this limitation, the proposed framework integrates multiple fine-tuned transformerbased summarization models and introduces a metric-based selection mechanism. In this framework, each model independently generates a candidate summary for the same input article. The generated summaries are then evaluated using automatic evaluation metrics that capture both lexical similarity and semantic relevance. Based on these scores, the framework selects the highest-quality summary as the final output. The models are finetuned and evaluated on the widely used CNN/DailyMail news summarization dataset. Experimental results demonstrate that the proposed framework achieves the highest BERTScore among all compared methods with a score of 88.63% . It also outperforms several LLMs such as GPT3-D2, Falcon-7b, and Mpt-7b, highlighting its effectiveness and robustness. These findings highlight the effectiveness of leveraging multiple transformer-based models within a metric-based selection strategy to improve the quality and robustness of automatic text summarization systems.  \nIndex Terms—Abstractive summarization, multi-model framework, metric-based selection, large language models, automatic text summarization.  \nI. INTRODUCTION  \nAs a consequence of rapid growth of digital data, there is a tremendous amount of texts available through different media sources, social networks, and Internet databases. The ability to quickly analyze and comprehend the text is becoming a more and more pressing issue. Text summarization has become a key task for NLP, which concentrates on creating compact summaries that contain all the important information contained in the original text. News summarization is one of the most important tasks since it helps readers to get key information from long articles [1], [2] .  \nTraditional methods used in this field have mostly relied on extractive systems which focus on selecting the most useful sentences from the text. Such methods usually follow a sequence of steps including text processing, feature  \n979-8-3315-8488-7/26/$31.00 ©2026 IEEE  \nextraction, scoring of sentences, applying a base model, sentence selection, and generating the summary. An analysis of various methods of extractive summarization will show the variety of methods used such as statistical methods, rulebased methods, fuzzy logic, optimization methods, graphbased, clustering-based, machine learning and deep learning methods [1] . Although certain extraction techniques have helped greatly in developing summarizers, many studies have shown that today’s systems still encounter difficulties which include lack of robustness on different articles’ structure and subject matter, poor performance when employing only one model architecture, and dependency on a single model output that does not always address the lexical and semantic aspects of good summaries [1], [3], [4] .  \nTransformers have demonstrated impressive performance on benchmark tasks like CNN/DailyMail, due to their selfattention mechanism and ability to process sequences in parallel, which help in capturing long-distance textual relationships [5], [6] . In the same vein, researchers have proposed hybrid approaches for OCR and summarizat","cbCailZKiHYXWBQU","https://ap.wps.com/l/cbCailZKiHYXWBQU","pdf",1969180,1,6,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"What problem does the paper address in abstractive text summarization?\",\"answer\":\"Single-model abstractive summarization can produce inconsistent quality across articles with different structures and topics, and may not balance lexical overlap with semantic relevance effectively.\"},{\"question\":\"How does the proposed framework generate the final summary?\",\"answer\":\"Multiple fine-tuned transformer models each generate a candidate summary for the same input article, then automatic evaluation metrics score lexical similarity and semantic relevance to rank candidates and select the best one.\"},{\"question\":\"Which dataset and evaluation results are reported?\",\"answer\":\"The framework is fine-tuned and evaluated on the CNN/DailyMail news summarization dataset, achieving the highest BERTScore of 88.63% and outperforming several LLM baselines mentioned in the paper.\"}]",1784204894,15,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"a-multi-model-metric-based-selection-framework-for-abstractive-text-summarization","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/a-multi-model-metric-based-selection-framework-for-abstractive-text-summarization/85608/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-25","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the paper address in abstractive text summarization?","Question",{"text":75,"@type":76},"Single-model abstractive summarization can produce inconsistent quality across articles with different structures and topics, and may not balance lexical overlap with semantic relevance effectively.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed framework generate the final summary?",{"text":80,"@type":76},"Multiple fine-tuned transformer models each generate a candidate summary for the same input article, then automatic evaluation metrics score lexical similarity and semantic relevance to rank candidates and select the best one.",{"name":82,"@type":73,"acceptedAnswer":83},"Which dataset and evaluation results are reported?",{"text":84,"@type":76},"The framework is fine-tuned and evaluated on the CNN/DailyMail news summarization dataset, achieving the highest BERTScore of 88.63% and outperforming several LLM baselines mentioned in the paper.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]