[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-327485-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-327485-en":130},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","neuron-aware-active-few-shot-learning-for-llms","Neuron-Aware Active Few-Shot Learning for LLMs","","Active Few-Shot Learning (AFSL) adapts large language models to specialized domains by selecting high-value unlabeled samples for annotation and turning them into few-shot demonstrations, aiming to cut human labeling costs while improving task performance. Existing AFSL techniques often depend on output-level signals such as entropy or external-embedding similarities, which can miss the model’s internal knowledge gaps and dynamics. NEUFS replaces proxy signals with neuron activation patterns and uses a dual-criteria strategy balancing diversity and hallucination-prone sample selection via neuron consensus. Experiments on three datasets validate stronger reasoning and classification results, and ablation studies show internal neuron activations provide a more principled selection signal than external embeddings.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/neuron-aware-active-few-shot-learning-for-llms/327485/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/neuron-aware-active-few-shot-learning-for-llms/327485.png","ImageObject",300,407,{"name":42,"@type":43},"Hazel","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-28","2026-09-21",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",9,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What problem does NEUFS address in active few-shot learning for LLMs?","Question",{"text":63,"@type":64},"NEUFS addresses the reliance of existing AFSL methods on output-level proxies that can overlook the model’s internal knowledge gaps and dynamics when selecting unlabeled samples for annotation.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How does NEUFS select few-shot samples differently from prior work?",{"text":68,"@type":64},"NEUFS uses neuron activation patterns from the model as direct representations of samples, then applies a dual-criteria strategy to ensure diversity and to prioritize informative, hallucination-prone examples.",{"name":70,"@type":61,"acceptedAnswer":71},"What evidence supports NEUFS effectiveness?",{"text":72,"@type":64},"Experiments on three datasets show improved performance on reasoning and text classification tasks, and ablation studies indicate neuron activations are a more principled and effective selection signal than external embeddings.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},327485,1790717856,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,119,122,126],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":55,"doc_module":4,"doc_module_name":25,"category_name":116,"show_sort_weight":117,"slug":118},"Religion & Spirituality",20,"religion-spirituality",{"id":117,"doc_module":4,"doc_module_name":25,"category_name":120,"show_sort_weight":117,"slug":121},"World Cup","world-cup",{"id":123,"doc_module":4,"doc_module_name":25,"category_name":124,"show_sort_weight":123,"slug":125},10,"Lifestyle","lifestyle",{"id":127,"doc_module":4,"doc_module_name":25,"category_name":128,"show_sort_weight":97,"slug":129},19,"General","general",{"code":4,"msg":82,"data":131},{"doc_id":79,"user_id":132,"nickname":42,"user_avatar":133,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":139,"language":140,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":12,"update_tm":144,"read_time":109,"preview_page_count":4},137441390410,"https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984","Neuron-Aware Active Few-Shot Learning for LLMs  \nZhuowei Chen Liwei Chen Christian Schunn Raquel Coelho Xiang Lorraine Li  \nUniversity of Pittsburgh, USA {zhuowei.chen, [xianglli}@pitt.edu](xianglli}@pitt.edu)  \narXiv :2607 .02423v 1 [ cs .LG] 2 Jul 2026  \nAbstract  \nActive Few-Shot Learning (AFSL) adapts LLMs to specialized domains by identifying the most valuable unlabeled samples for annotation and use as few-shot demonstrations, effectively reducing human annotation costs while promoting high performance. However, existing methods typically rely on output-level signals for the sample identification, such as predictive entropy or semantic similarities with test-time data based on external embeddings, which often overlook models’ internal dynamics which could pinpoint specific knowledge gaps. To bridge this gap, we propose NEUFS, a Neuron-Aware Active Few-Shot Learning framework that shifts the selection paradigm from output-level proxies to models’ internal dynamics. NEUFS utilizes neuron activation patterns to represent sample directly, and includes a dual-criteria selection strategy that: (1) ensures few-shot sample diversity with neuron patterns for broader example coverage, while  \n(2) prioritizing on identifying informative and challenging few-shot samples LLMs tend to hallucinate by quantifying neuron consensus. Experiments on three datasets demonstrate that NEUFS excels in both reasoning and text classification tasks, outperforming existing AFSL baselines. Ablation studies further highlight that internal neuron activations provide a more principled and effective selection signal than external embeddings, validating the superiority of the proposed NEUFS.  \n1 Introduction  \nWhile recent advances in LLMs have greatly improved general tasks like text classification, these models often fail to perform well out-of-the-box in specialized domains such as education, medicine, or law. Adapting off-the-shelf models to these expert fields is typically made difficult by the scarcity of data and the computational costs required for fine-tuning. To adapt LLMs for specific downstream tasks, few-shot in-context learning (ICL)  \nFigure 1: Neuron-Aware AFSL, which identifies informative and diverse few-shot samples for annotation. By leveraging activation patterns, NEUFS ensures broad knowledge coverage through activation pattern-based clustering while targeting informative, hallucinationprone samples via neuron consensus quantification.  \nhas emerged as a popular training-free adaptation strategy. By providing examples within the query context, ICL has demonstrated promising performance in practical applications (Brown et al., 2020) . The selection of few-shot has been identified as a critical factor for downstream task performance (Min et al., 2022) .  \nExisting work on few-shot example selection typically assumes access to a large annotated dataset, where retrieving examples most similar to the test data at inference time yields the best performance (Yu et al., 2023a ; Margatina et al., 2023) . However, constructing large annotated datasets for test-time retrieval is oftentimes impractical, especially for specialized domains that require expensive expert-level annotations. Consequently, exploration of few-shot selection strategies under unlabeled data settings remains limited. Recent work has begun to address this gap by introducing active learning (AL) into few-shot ICL, i.e Active FewShot Learning (AFSL) (Ahmadnia et al., 2025) .  \nAFSL extends traditional AL techniques to the few-shot paradigm, selecting the samples from a large unlabeled pool that are most likely to yield strong test-time performance, to be annotated as ICL demonstrations, providing a promising solu-  \ntion for reducing annotation costs while maintaining performance (Xia et al., 2025 ; Ahmadnia et al., 2025) . AFSL methods largely follow the intuitionsof mainstream AL approaches, which either prioritize highly uncertain samples to select informative samples (Men","cbCaihPTApuECSs6","https://ap.wps.com/l/cbCaihPTApuECSs6","pdf",1616737,16,"English","# Introduction\n## Active few-shot learning and domain adaptation\n## Output-level selection signals and their limitations\n## NEUFS: neuron-aware internal dynamics for selection","[{\"question\":\"What problem does NEUFS address in active few-shot learning for LLMs?\",\"answer\":\"NEUFS addresses the reliance of existing AFSL methods on output-level proxies that can overlook the model’s internal knowledge gaps and dynamics when selecting unlabeled samples for annotation.\"},{\"question\":\"How does NEUFS select few-shot samples differently from prior work?\",\"answer\":\"NEUFS uses neuron activation patterns from the model as direct representations of samples, then applies a dual-criteria strategy to ensure diversity and to prioritize informative, hallucination-prone examples.\"},{\"question\":\"What evidence supports NEUFS effectiveness?\",\"answer\":\"Experiments on three datasets show improved performance on reasoning and text classification tasks, and ablation studies indicate neuron activations are a more principled and effective selection signal than external embeddings.\"}]","Neuron-Aware Active Few-Shot Learning for LLMs | PDF",1789977307]