[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-81813-en":3,"doc-seo-81813-105":31,"detail-sidebar-cat-0-en-105":93},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},81813,137441390410,"Hazel","https://ap-avatar.wpscdn.com/avatar/2000252f4ab5702993?_k=1776741390130283984",8,"Research & Report","CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse","Divergent thinking drives creativity, yet large language models often produce repetitive answers to open-ended questions, known as the artificial hivemind effect. CreativityNeuro presents a data-free contrastive weight steering approach that enhances divergent thinking in LLMs. Evaluations across multiple creativity assessments show gains up to 14 human percentile points on DAT and significant improvements in originality, surprise, and creativity on AUT and TT, while consistently reducing mode collapse.","arXiv :2607 .0 1433v 1 [ cs .AI] 1 Jul 2026  \nCreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse  \nSamuel Schapiro∗  \nUniveristy of Illinois, Urbana-Champaign  \nCore Francisco Park  \nCenter for Brain Science, Harvard University  \nCBS-NTT Program in Physics of Intelligence, Harvard University Prior Computers  \nFelix Sosa  \nPrior Computers  \nLav R. Varshney  \nAI Innovation Institute, Stony Brook University  \nFigure 1: CreativityNeuro (CN) pipeline. Given a pair of contrastive creative prompts, CN computes parameter importance scores, selects a sparse subset of creativity-relevant parameters, and applies a scaled weight perturbation—without requiring behavioral datasets or gradient-based finetuning. CN improves divergent thinking across various tasks. Subplot (b) visualizes CN thinking outside of the “box”(i.e., the convex hull of baseline DAT responses), despite baseline responses falling within CN’s convex hull in subplot (a) .  \nAbstract  \nDivergent thinking is a crucial aspect of creativity, yet large language models (LLMs) tend to consistently generate similar responses to open-ended questions, in what has been termed the artificial hivemind effect. Here, we introduce CreativityNeuro, a data-free method for enhancing divergent thinking in LLMs via contrastive weight steering. We evaluate our method across multiple creativity assessments and report several main findings.  \nOn the Divergent Association Task (DAT), a vocabulary-space creativity  \n∗ Corresponding author: [schapironietzsche@gmail.com](schapironietzsche@gmail.com)  \ntest, CreativityNeuro improves performance by up to 14 human percentile points. Next, in a large-scale human evaluation (N=720) on the Alternative Uses Test (AUT) and the Task Task, CreativityNeuro achieves significant improvements in originality, surprise, and creativity, transferring to longerform and more open-ended tasks. Importantly, we find that across all three tasks, CreativityNeuro demonstrably reduces measures of mode collapse.  \nMoreover, activation steering achieves comparable performance to CreativityNeuro on the DAT, but it does not transfer to the AUT and Task Task, demonstrating the effectiveness of weight-space steering in generalizing to unseen tasks. In conclusion, CreativityNeuro improves divergent thinking and reduces mode collapse without requiring behavioral data, re-training, or gradient-based fine-tuning, providing a straightforward way to enhance LLM performance in creative domains.  \n1 Introduction  \nRecent advances in large language models (LLMs) have renewed interest in a longstanding question: how can we understand and enhance creativity in intelligent systems? (Boden, 2004) . While this question has deep roots in cognitive science (Quetelet, 1842; Galton, 1870; Hadamard, 1954; Guilford, 1956; Mednick, 1962; Koestler, 1964; Simonton, 2004; Dietrich, Arne, 2004; Fauconnier & Turner, 2008; Rothenberg, 2014), it is now increasingly studied in the context of large-scale generative models (Maher, 2010; Varshney, 2019; Schapiro et al., 2025) . Recent work has begun to assess the capacity for LLMs to engage in creative and open-ended tasks (Si et al., 2024; 2025; Sanyal et al., 2025; Bellemare-Pepin et al., 2024; Wanget al., 2025; 2024), where a recurring issue has surfaced: models tend to consistently generate similar responses to open-ended questions, in what has been termed the artificial hivemind effect (Jiang et al., 2025) .  \nWithin the creativity literature, a common distinction is made between divergent thinking, the capacity to generate multiple diverse solutions to a problem, and convergent thinking, the ability to find a single correct solution that unifies multiple diverse stimuli (Dietrich, 2019; Guilford, 1956) . Studying ways to enhance divergent thinking offers a promising pathway to encourage greater diversity and novelty in model responses, combating the homogenization issues that have emerged thus far. Here, ","cbCaitborD69HWMg","https://ap.wps.com/l/cbCaitborD69HWMg","pdf",8180293,5,1,27,"English","en",105,"# Abstract\n# Introduction\n# Related Work","[{\"question\":\"What problem does CreativityNeuro address in large language models?\",\"answer\":\"CreativityNeuro targets the tendency of LLMs to generate similar responses to open-ended questions, referred to as the artificial hivemind effect, and it aims to reduce mode collapse.\"},{\"question\":\"How does CreativityNeuro improve divergent thinking without using behavioral data or gradient-based fine-tuning?\",\"answer\":\"CreativityNeuro performs contrastive weight steering by computing parameter importance scores from creative vs. non-creative prompts, selecting sparse creativity-relevant parameters, and applying a scaled weight perturbation.\"},{\"question\":\"How does CreativityNeuro compare with activation steering on different creativity tasks?\",\"answer\":\"CreativityNeuro achieves strong improvements on DAT, AUT, and TT and reduces mode collapse across all three tasks, while activation steering performs comparably on DAT but transfers poorly to AUT and TT.\"}]","CreativityNeuro: Steering Language Model Weights to Improve Divergent Thinking and Reduce Mode Collapse | PDF",1784176316,68,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":88,"head_meta":90,"extra_data":92,"updated_unix":29},"creativityneuro-steering-language-model-weights-to-improve-divergent-thinking-and-reduce-mode-collapse","",{"@graph":37,"@context":87},[38,55,70],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,52],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":51},"https://docshare.wps.com/document/research-report/",3,{"item":53,"name":13,"@type":44,"position":54},"https://docshare.wps.com/document/creativityneuro-steering-language-model-weights-to-improve-divergent-thinking-and-reduce-mode-collapse/81813/",4,{"url":53,"name":13,"@type":56,"author":57,"headline":13,"publisher":59,"fileFormat":62,"inLanguage":24,"description":14,"dateModified":63,"datePublished":64,"encodingFormat":62,"isAccessibleForFree":65,"interactionStatistic":66},"DigitalDocument",{"name":9,"@type":58},"Person",{"url":42,"name":60,"@type":61},"DocShare","Organization","application/pdf","2026-07-30","2026-07-16",true,{"@type":67,"interactionType":68,"userInteractionCount":20},"InteractionCounter",{"@type":69},"ViewAction",{"@type":71,"mainEntity":72},"FAQPage",[73,79,83],{"name":74,"@type":75,"acceptedAnswer":76},"What problem does CreativityNeuro address in large language models?","Question",{"text":77,"@type":78},"CreativityNeuro targets the tendency of LLMs to generate similar responses to open-ended questions, referred to as the artificial hivemind effect, and it aims to reduce mode collapse.","Answer",{"name":80,"@type":75,"acceptedAnswer":81},"How does CreativityNeuro improve divergent thinking without using behavioral data or gradient-based fine-tuning?",{"text":82,"@type":78},"CreativityNeuro performs contrastive weight steering by computing parameter importance scores from creative vs. non-creative prompts, selecting sparse creativity-relevant parameters, and applying a scaled weight perturbation.",{"name":84,"@type":75,"acceptedAnswer":85},"How does CreativityNeuro compare with activation steering on different creativity tasks?",{"text":86,"@type":78},"CreativityNeuro achieves strong improvements on DAT, AUT, and TT and reduces mode collapse across all three tasks, while activation steering performs comparably on DAT but transfers poorly to AUT and TT.","https://schema.org",{"og:url":53,"og:type":89,"og:title":13,"og:site_name":60,"og:description":14},"article",{"robots":91,"canonical":53},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":94},[95,99,103,107,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":96,"show_sort_weight":97,"slug":98},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":100,"show_sort_weight":101,"slug":102},"Literature",80,"literature",{"id":54,"doc_module":4,"doc_module_name":47,"category_name":104,"show_sort_weight":105,"slug":106},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":47,"category_name":138,"show_sort_weight":20,"slug":139},19,"General","general"]