[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-1-en-105":3,"doc-seo-239146-105":53,"doc-detail-239146-en":126},{"code":4,"msg":5,"data":6},0,"success",[7,14,19,24,29,34,39,44,49],{"id":8,"doc_module":9,"doc_module_name":10,"category_name":11,"show_sort_weight":12,"slug":13},11,1,"Template","Presentations",90,"presentations",{"id":15,"doc_module":9,"doc_module_name":10,"category_name":16,"show_sort_weight":17,"slug":18},12,"Resumes",80,"resumes",{"id":20,"doc_module":9,"doc_module_name":10,"category_name":21,"show_sort_weight":22,"slug":23},14,"Invoices",70,"invoices",{"id":25,"doc_module":9,"doc_module_name":10,"category_name":26,"show_sort_weight":27,"slug":28},15,"Posters",60,"posters",{"id":30,"doc_module":9,"doc_module_name":10,"category_name":31,"show_sort_weight":32,"slug":33},16,"Social Media",50,"social-media",{"id":35,"doc_module":9,"doc_module_name":10,"category_name":36,"show_sort_weight":37,"slug":38},17,"Forms",40,"forms",{"id":40,"doc_module":9,"doc_module_name":10,"category_name":41,"show_sort_weight":42,"slug":43},18,"Letters",30,"letters",{"id":45,"doc_module":9,"doc_module_name":10,"category_name":46,"show_sort_weight":47,"slug":48},21,"Paper Templates",5,"papers-templates",{"id":50,"doc_module":9,"doc_module_name":10,"category_name":51,"show_sort_weight":4,"slug":52},158,"General","general-158",{"code":4,"msg":54,"data":55},"ok",{"site_id":56,"language":57,"slug":58,"title":59,"keywords":60,"description":61,"schema_data":62,"social_meta":119,"head_meta":121,"extra_data":123,"updated_unix":125},105,"en","template-free-prompt-tuning-for-few-shot-ner","Template-free Prompt Tuning for Few-shot NER","","Prompt-based methods work well in sentence-level few-shot learning, largely due to carefully designed templates and label words. For token-level tasks like NER, template enumeration over all candidate entity spans becomes time-consuming and inefficient. Template-free Prompt Tuning reformulates NER as a language-modeling problem without templates by predicting a class-related pivot or label word at entity positions. Automatic label-word search further reduces the pretrain–finetune objective gap and improves few-shot performance, yielding large decoding speed gains.",{"@graph":63,"@context":118},[64,80,101],{"@type":65,"itemListElement":66},"BreadcrumbList",[67,71,74,77],{"item":68,"name":69,"@type":70,"position":9},"https://docshare.wps.com","Home","ListItem",{"item":72,"name":10,"@type":70,"position":73},"https://docshare.wps.com/template/",2,{"item":75,"name":51,"@type":70,"position":76},"https://docshare.wps.com/template/general/",3,{"item":78,"name":59,"@type":70,"position":79},"https://docshare.wps.com/template/template-free-prompt-tuning-for-few-shot-ner/239146/",4,{"url":78,"name":59,"@type":81,"image":82,"author":87,"headline":59,"publisher":90,"fileFormat":93,"inLanguage":57,"description":61,"dateModified":94,"datePublished":95,"encodingFormat":93,"isAccessibleForFree":96,"interactionStatistic":97},"DigitalDocument",{"url":83,"@type":84,"width":85,"height":86},"https://docshare.wps.com/thumbnails/template-free-prompt-tuning-for-few-shot-ner/239146.png","ImageObject",442,249,{"name":88,"@type":89},"Aditya","Person",{"url":68,"name":91,"@type":92},"DocShare","Organization","application/pdf","2026-09-23","2026-09-11",true,{"@type":98,"interactionType":99,"userInteractionCount":79},"InteractionCounter",{"@type":100},"ViewAction",{"@type":102,"mainEntity":103},"FAQPage",[104,110,114],{"name":105,"@type":106,"acceptedAnswer":107},"Why are template-based prompts inefficient for NER in few-shot settings?","Question",{"text":108,"@type":109},"They require enumerating templates and querying over all possible spans for token labels, which becomes time-consuming and grows rapidly with sentence length.","Answer",{"name":111,"@type":106,"acceptedAnswer":112},"How does the proposed method reformulate NER without templates?",{"text":113,"@type":109},"It reformulates NER as an LM task and fine-tunes the model to predict class-related pivot (label) words at entity positions while keeping original-word prediction for non-entity positions.",{"name":115,"@type":106,"acceptedAnswer":116},"What is the role of automatically searching label words?",{"text":117,"@type":109},"It selects or derives label words that the pretrained model can adapt to effectively, helping reduce the objective gap between pretraining and fine-tuning and improving few-shot results.","https://schema.org",{"og:url":78,"og:type":120,"og:title":59,"og:site_name":91,"og:description":61},"article",{"robots":122,"canonical":78},"index,follow",{"doc_id":124,"site_id":56},239146,1790045038,{"code":4,"msg":5,"data":127},{"doc_id":124,"user_id":128,"nickname":88,"user_avatar":129,"doc_module":9,"category_id":50,"category_name":51,"doc_title":59,"doc_description":61,"doc_content":130,"file_id":131,"file_url":132,"file_type":133,"file_size":134,"view_count":79,"is_deleted":4,"is_public":9,"is_downloadable":9,"audit_status":9,"page_count":15,"language":135,"language_code":57,"site_id":56,"html_lang":57,"table_of_contents":136,"faqs":137,"seo_title":138,"seo_description":61,"update_tm":139,"read_time":79},962085564549,"https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45","Template-free Prompt Tuning for Few-shot NER  \nRuotian Ma 1∗, Xin Zhou 1∗, Tao Gui2†, Yiding Tan 1 , Linyang Li 1 , Qi Zhang 1†, Xuanjing Huang 1  \n1 School of Computer Science, Fudan University, Shanghai, China  \n2Institute of Modern Languages and Linguistics, Fudan University, Shanghai, China {rtma19,xzhou20,tgui,qz,[xjhuang}@fudan.edu.cn](xjhuang}@fudan.edu.cn)  \nAbstract  \nPrompt-based methods have been successfully applied in sentence-level few-shot learning tasks, mostly owing to the sophisticated design of templates and label words. However, when applied to token-level labeling tasks such as NER, it would be time-consuming to enumerate the template queries over all potential entity spans. In this work, we propose a more elegant method to reformulate NER tasks as LM problems without any templates. Specifically, we discard the template construction process while maintaining the word prediction paradigm of pre-training models to predict a class-related pivot word (or label word) at the entity position. Meanwhile, we also explore principled ways to automatically search for appropriate label words that the pre-trained models can easily adapt to. While avoiding the complicated template-based process, the proposed LM objective also reduces the gap between different objectives used in pre-training and fine-tuning, thus it can better benefit the few-shot performance. Experimental results demonstrate the effectiveness of the proposed method over bert-tagger and template-based method under few-shot settings. Moreover, the decoding speed of the proposed method is up to 1930.12 times faster than the template-based method.  \n1 Introduction  \nPre-trained language models (LMs) have led to large improvements in NLP tasks (Devlin et al., 2019 ; Liu et al., 2019 ; Lewis et al., 2020) . Popular practice to perform downstream classification tasks is to replace the pretrained model’s output layer with a classifier head and fine-tune it using atask-specific objective function. Recently, a new paradigm, prompt-based learning, has achieved great success on few-shot classification tasks by reformulating classification tasks as cloze  \n∗ Equal contribution.  \n† Corresponding authors.  \nLM predictions  \n| [CLS] Input  Obama is a [MASK] entity. [SEP]\u003Cbr>|\n| --- |\n| [CLS] Input  Obama was is a [MASK] entity. [SEP]\u003Cbr>|\n| ...... |\n| [CLS] Input America . is a [MASK] entity. [SEP]\u003Cbr>|\n\n Query LM x 1  \n Query LM x 2  \n Query LM x 21  \nFigure 1: An example of template-based prompt method for NER. Predicting all labels in sentence “Obama was born in America.\" requires enumeration over all spans.  \nquestions. Typically, for each input [X], a template is used to convert [X] into an unfilled text (e.g.,“[X] It was   .\"), allowing the model to fill in the blank with its language modeling ability. For instance, when performing sentiment classification task, the input “I love the milk.\" can be converted into “I love the milk. It was   . \" . Consequently, the LM may predict a label word “great\", indicating that the input belongs to a positive class.  \nTwo main factors contribute to the success of prompt-based learning on few-shot classification. First, re-using the masked LM objective helps alleviate the gap between different training objectives used at pre-training and fine-tuning. Therefore, the LMs can faster adapt to downstream tasks even with a few training samples (Schick and Schütze, 2021a,b ; Brown et al., 2020) . Second, the sophisticated template and label word design helps LMs better fit the task-specific answer distributions, which also benefits few-shot performance. As proved in previous works, proper templates designed by manually selecting (Schick and Schütze, 2021a,b), gradient-based discrete searching (Shin et al., 2020), LM generating (Gao et al., 2021) and continuously optimizing (Liu et al., 2021) are able to induce the LMs to predict more appropriate answers needed in corresponding tasks. However, the template-based prompt methods are intr","cbCaio2OusiXM1XB","https://ap.wps.com/l/cbCaio2OusiXM1XB","pdf",485641,"English","# Introduction\n## Template-based prompt learning for NER\n## Proposed template-free EntLM objective","[{\"question\":\"Why are template-based prompts inefficient for NER in few-shot settings?\",\"answer\":\"They require enumerating templates and querying over all possible spans for token labels, which becomes time-consuming and grows rapidly with sentence length.\"},{\"question\":\"How does the proposed method reformulate NER without templates?\",\"answer\":\"It reformulates NER as an LM task and fine-tunes the model to predict class-related pivot (label) words at entity positions while keeping original-word prediction for non-entity positions.\"},{\"question\":\"What is the role of automatically searching label words?\",\"answer\":\"It selects or derives label words that the pretrained model can adapt to effectively, helping reduce the objective gap between pretraining and fine-tuning and improving few-shot results.\"}]","Template-free Prompt Tuning for Few-shot NER | PDF",1789144515]