[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-267355-105":59,"doc-detail-267355-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","can-language-models-learn-analogical-reasoning-investigating-training-objectives-and-comparisons-to-human-performance","Can Language Models Learn Analogical Reasoning? Investigating Training Objectives and Comparisons to Human Performance","","Analogies are commonly used to evaluate word embeddings in NLP, yet the question remains whether analogical reasoning itself can be learned as a standalone capability. This paper tests multiple training approaches for learning basic analogical reasoning, emphasizing analogy types that better reflect those used in human psychometric assessments rather than standard NLP benchmarks. Results show that models can learn analogical reasoning with limited data, and that after training their performance approaches a human baseline.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/can-language-models-learn-analogical-reasoning-investigating-training-objectives-and-comparisons-to-human-performance/267355/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/can-language-models-learn-analogical-reasoning-investigating-training-objectives-and-comparisons-to-human-performance/267355.png","ImageObject",300,407,{"name":92,"@type":93},"McGucket","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-29","2026-09-14",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",16,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does the paper address about analogies in NLP?","Question",{"text":113,"@type":114},"The paper investigates whether analogical reasoning can be learned, not just used as an evaluation signal for word embeddings. It focuses on training models with analogy types closer to those used in human tests.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the paper’s evaluation differ from common NLP analogy benchmarks?",{"text":118,"@type":114},"It emphasizes analogies that resemble human psychometric analogy tests instead of the morphological and encyclopedic relations common in benchmarks like Google Analogy Test Set and BATS.",{"name":120,"@type":111,"acceptedAnswer":121},"What are the main findings after training?",{"text":122,"@type":114},"The experiments show that models learn analogical reasoning even with a small amount of data, and their performance after training approaches a human baseline on an unseen human-focused test set.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},267355,1789417529,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":145,"preview_page_count":4},1236954412713,"https://us-avatar.wpscdn.com/davatar_29158cc5080c5b710cf443261637dec0","Can Language Models Learn Analogical Reasoning? Investigating Training Objectives and Comparisons to Human Performance  \nMolly R. Petersen1,2 and Lonneke van der Plas1,3  \n1 Computation, Cognition & Language Group, Idiap Research Institute, Martigny, Switzerland  \n2 NLP Lab, EPFL, Lausanne, Switzerland  \n3 Institute of Linguistics and Language Technology, University of Malta, Malta  \n[molly.petersen@idiap.ch](molly.petersen@idiap.ch), [lonneke.vanderplas@idiap.ch](lonneke.vanderplas@idiap.ch)  \nAbstract  \nWhile analogies are a common way to evaluate word embeddings in NLP, it is also of interest to investigate whether or not analogical reasoning is a task in itself that can be learned. In this paper, we test several ways to learn basic analogical reasoning, specifically focusing on analogies that are more typical of what is used to evaluate analogical reasoning in humans than those in commonly used NLP benchmarks. Our experiments find that models are able to learn analogical reasoning, even with a small amount of data. We additionally compare our models to a dataset with a human baseline, and find that after training, models approach human performance.  \n1 Introduction  \nSolving proportional analogies a : b :: c : d (ex. Paris:France::Tokyo:Japan) (Mikolov et al., 2013a,b ; Rogers et al., 2017) with embeddings has become an iconic way to demonstrate that they encode semantic information regarding the relationships between word pairs. Proportional analogies have been formalized with word embeddings by means of a simple arithmetic equation (often referred to as the vector offset method) b − a+c = d where the choice of d is solved by maximizing cos (b−a+c, d), or other slightly more complicated similarity measures (Mikolov et al., 2013a ; Rogerset al., 2017 ; Levy and Goldberg, 2014 ; Ushio et al., 2021b) .  \nThis task has developed into many datasets such as the Google Analogy Test Set and the Bigger Analogy Test Set (BATS) (Mikolov et al., 2013a ; Gladkova et al., 2016) . Examples of relation types included in these datasets are morphological in nature (entertain : entertainer :: examine : examiner), based on encyclopedic knowledge (Paris : France :: Lima : Peru), or lexicographic relations, such as hypernyms and hyponyms (candy : lollipop :: color : white) . Notably, the relations between entities that form each  \nanalogy are explicitly verbalized for each example, and are grouped into collections of many pairs with equivalent relations. While the ability of word embeddings to solve these sorts of analogies is interesting, the analogies contained in these datasets are different from what is typically used to test analogical reasoning in humans. Many would be trivially easy to solve, as is the case with morphological relations and a lot of encyclopedic knowledge (Ushio et al., 2021b) . Even the task of completing an analogy by filling in a correct d term is perhaps not well suited to humans. Rogers et al. (2017) noted that analogy questions are often given to people as multiple choice questions, as people would likely fill in d with a variety of words where a single correct answer is not always the case. It has also been pointed out that model performance on analogy tests tend to rely on how semantically related all the entities in the analogies are to each other, suggesting these tests might in fact not be measuring analogical reasoning (Rogers et al., 2017) . When using the vector offset method, the a, b, and c terms are generally excluded from candidates for d. If they are not excluded b or c is often the closest neighbor to the estimate of d and will be selected (Rogers et al., 2017 ; Linzen, 2016) .  \nAnalogical reasoning has the potential to extend beyond just solving proportional analogies. Analogies and analogical reasoning have value regarding scientific discovery, problem solving, human creative endeavors such as metaphor generation, as well as in the field of education (Boger et al., 2019 ; Gentner, 2002 ; Gentner et","cbCailcI91WZTdy5","https://ap.wps.com/l/cbCailcI91WZTdy5","pdf",262556,18,"English","# Introduction\n## Analogies in NLP and word embeddings\n## Motivating psychometric-style analogy tests\n## Contributions and training objective\n# Related Work\n## Datasets designed for human analogy reasoning\n## Scoring methods and difficulty comparison","[{\"question\":\"What problem does the paper address about analogies in NLP?\",\"answer\":\"The paper investigates whether analogical reasoning can be learned, not just used as an evaluation signal for word embeddings. It focuses on training models with analogy types closer to those used in human tests.\"},{\"question\":\"How does the paper’s evaluation differ from common NLP analogy benchmarks?\",\"answer\":\"It emphasizes analogies that resemble human psychometric analogy tests instead of the morphological and encyclopedic relations common in benchmarks like Google Analogy Test Set and BATS.\"},{\"question\":\"What are the main findings after training?\",\"answer\":\"The experiments show that models learn analogical reasoning even with a small amount of data, and their performance after training approaches a human baseline on an unseen human-focused test set.\"}]","Can Language Models Learn Analogical Reasoning? Investigating Training Objectives and Comparisons to Human Performance | PDF",45]