[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84827-en":3,"doc-seo-84827-105":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},84827,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","Text Distance from Nested and Hierarchical Repetitions: A Compression-Based Perspective","A new structural sequence analysis method grounded in Algorithmic Information Theory (AIT) introduces the Ladderpath approach to extract nested and hierarchical relationships among repeated substructures in linguistic sequences. The extracted structures define three text-distance measures: a normalized compression distance (NCD) and two additional distances derived from the Ladderpath representation. Coupled with a k-nearest neighbor classifier, the distances deliver strong, consistent text classification performance across in-distribution, out-of-distribution, and few-shot settings, outperforming gzip-based NCD and BERT under OOD and low-resource regimes.","arXiv :2607 .05416v1 [ cs .CL] 23 Jun 2026  \nTEXT DISTANCE FROM NESTED AND HIERARCHICAL REPETITIONS: A COMPRESSION-BASED PERSPECTIVE  \nXiaojun Hu1 ,2 ,3 , Jing Wang1 ,2 ,3 , Jingwen Zhang1 ,2 , Fengyao Zhai1 ,2 ,3 , Xiao Xie1 ,4 , Hao Liao5 , Zengru Di1 ,2 , Yu Liu1 ,2 ,†  \n1Department of Systems Science, Faculty of Arts and Sciences, Beijing Normal University, Zhuhai, China.  \n2International Academic Center of Complex Systems, Beijing Normal University, Zhuhai, China.  \n3 School of Systems Science, Beijing Normal University, Beijing, China.  \n4 School of Physics and Astronomy, Sun Yat-sen University, Zhuhai, China.  \n5 College of Computer Science and Software Engineering, Shenzhen University,Shenzhen,China.  \nABSTRACT  \nWe present a new method for structural sequence analysis grounded in Algorithmic Information Theory (AIT) . At its core is the Ladderpath approach, which extracts nested and hierarchical relationships among repeated substructures in linguistic sequences—an instantiation of AIT’s principle of describing data through minimal generative programs. These structures are then used to define three distance measures: a normalized compression distance (NCD), and two alternative distances derived directly from the Ladderpath representation. Integrated with a k-nearest neighbor classifier, these distances achieve strong and consistent performance across in-distribution, out-of-distribution (OOD), and few-shot text classification tasks. In particular, all three methods outperform both gzip-based NCD and BERT under OOD and low-resource settings. These results demonstrate that the structured representations captured by Ladderpath preserve intrinsic properties of sequences and provide a lightweight, interpretable, and training-free alternative for text modeling. This work highlights the potential of AIT-based approaches for structural and domain-agnostic sequence understanding.  \nKeywords Algorithmic Information Theory (AIT) · Normalized Compression Distance (NCD) · Compression · Ladderpath · Text Classification · Hierarchical Structure  \n1 Introduction  \nThe rapid advancements in natural language processing (NLP) and machine learning have significantly improved the performance of text classification and regression tasks [1, 2] . A wide range of approaches, from traditional bag-of-words models to deep neural architectures and pre-trained language models such as BERT [3], have been developed to tackle these tasks. However, deploying these models in low-resource or distributionally inconsistent scenarios remains a major challenge due to their heavy reliance on extensive annotated data and high computational demands [4, 5] . Although these models demonstrate good performance under ideal conditions, their generalization ability often diminishes in settings with scarce data or in the presence of domain shifts [6] .  \nTraditional models and word embedding-based classifiers require feature engineering or fine-tuning [7, 8], while largescale models encode a highly compressive mapping of data in an ultra-high-dimensional space. From an informationtheoretic perspective, the process of extracting patterns from data can be viewed as a form of compression—identifying and retaining only the most informative structures [9, 10] . Inspired by this insight, Jiang et al. proposed a parameter-free classification method that combines a standard compressor (e.g., gzip) with a k-nearest neighbor (k-NN) classifier to approximate deep learning-like performance without training [11, 12] . This method uses the normalized compression distance (NCD) to measure text similarity, offering a lightweight and generalized solution that is especially suited for low-resource and heterogeneous data. The theoretical foundation of this approach originates from the concept  \n∗These authors contributed equally.  \n†Corresponding author: [yu.ernest.liu@bnu.edu.cn](yu.ernest.liu@bnu.edu.cn)  \nof information distance and Kolmogorov complexity [13] . Bennett et al. introdu","cbCaimocMqn8MSWy","https://ap.wps.com/l/cbCaimocMqn8MSWy","pdf",2288537,1,19,"English","en",105,"# Introduction\n## Motivation and limitations of existing NLP models\n## Compression-based distance and training-free classification\n## Ladderpath-based structured compression approach","[{\"question\":\"What problem does the document address in text classification?\",\"answer\":\"It targets the difficulty of maintaining generalization in low-resource and distribution-shifted scenarios, where reliance on large annotated data and heavy computation can hurt performance.\"},{\"question\":\"How does the Ladderpath approach relate to Algorithmic Information Theory (AIT)?\",\"answer\":\"Ladderpath operationalizes an AIT principle by describing data through minimal generative programs, extracting nested and hierarchical repeated substructures in linguistic sequences.\"},{\"question\":\"How are the distance measures used for classification?\",\"answer\":\"Three compression-based distance measures—one normalized compression distance (NCD) and two derived from Ladderpath—are fed into a k-nearest neighbor classifier to perform training-free text classification.\"}]",1784198569,48,{"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},"text-distance-from-nested-and-hierarchical-repetitions-a-compression-based-perspective","",{"@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/text-distance-from-nested-and-hierarchical-repetitions-a-compression-based-perspective/84827/",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","2026-07-17","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 document address in text classification?","Question",{"text":75,"@type":76},"It targets the difficulty of maintaining generalization in low-resource and distribution-shifted scenarios, where reliance on large annotated data and heavy computation can hurt performance.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the Ladderpath approach relate to Algorithmic Information Theory (AIT)?",{"text":80,"@type":76},"Ladderpath operationalizes an AIT principle by describing data through minimal generative programs, extracting nested and hierarchical repeated substructures in linguistic sequences.",{"name":82,"@type":73,"acceptedAnswer":83},"How are the distance measures used for classification?",{"text":84,"@type":76},"Three compression-based distance measures—one normalized compression distance (NCD) and two derived from Ladderpath—are fed into a k-nearest neighbor classifier to perform training-free text classification.","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},[93,97,101,105,110,115,120,123,128,131,135],{"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":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":106,"slug":137},"General","general"]