[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86579-en":3,"doc-seo-86579-105":30,"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":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":13,"seo_description":14,"update_tm":28,"read_time":29},86579,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","CatRetriever: Contrastive Representation Learning for Slab-to-Bulk Retrieval in Generative Catalyst Discovery","Inverse design accelerates exploration of large chemical spaces to discover catalysts with targeted properties, while recent surface generative models directly synthesize catalyst surface–adsorbate structures. These models, however, generate only slab-level outputs and omit the parent bulk crystal, blocking reliable evaluation of formation energy, surface energy, crystallographic symmetry, and synthesizability. CatRetriever treats slab-to-bulk as a retrieval problem, learning shared latent representations and retrieving plausible bulk candidates with strong R@1 and R@3 scores. It further supports adsorption-energy targeted bulk discovery via retrieval, generative expansion, and adsorption-energy distribution analysis.","CatRetriever: Contrastive Representation Learning for Slab-to-Bulk Retrieval in Generative Catalyst Discovery  \nJungho Oh1,†, Woosung Kim2,†, Dong Hyeon Mok3, Jonggeol Na4,5,6* and Seoin Back1,6,7,8*  \nAUTHOR ADDRESS  \n1KU-KIST Graduate School of Converging Science and Technology, Korea University, Seoul 02841, Republic of Korea  \n2Department of Materials Science and Engineering, Korea University, Seoul 02841, Republic of Korea  \n3Department of Chemical and Biomolecular Engineering, Sogang University, Seoul 04107, Republic of Korea  \n4Department of Chemical Engineering and Materials Science, Ewha Womans University, Seoul 03760, Republic of Korea  \n5Department of Chemical Engineering, Graduate Program in System Health Science and Engineering, Ewha Womans University, Seoul, 03760, Republic of Korea  \n6Institute for Multiscale Matter and Systems (IMMS), Ewha Womans University, Seoul 03760, Republic of Korea  \n7Department of Integrative Energy Engineering, Korea University, Seoul 02841, Republic of Korea  \n8Center for Hydrogen and Fuel Cells, Korea Institute of Science and Technology(KIST), Seoul 02792, Republic of Korea  \n†These authors contributed equally to this work.  \n*Corresponding authors: [sback@korea.ac.kr](sback@korea.ac.kr) (SB), [jgna@ewha.ac.kr](jgna@ewha.ac.kr) (JN)  \nKEYWORDS  \nSlab-to-bulk retrieval, Contrastive learning, Adsorption energy-based screening, Catalyst discovery  \nABSTRACT  \nInverse design is an emerging data-driven paradigm for efficiently navigating vast chemical spaces to discover new materials with targeted properties, and in the context of heterogeneous catalysis, surface generative models have recently advanced this goal by directly generating catalyst surface-adsorbate structures. However, these models typically operate at the slab level and do not provide the corresponding parent bulk structure, making it difficult to assess bulkdependent properties such as formation energy, surface energy, crystallographic symmetry, and synthesizability. Here, we address this missing slab-to-bulk connection as a retrieval problem and introduce CatRetriever, a contrastive representation learning model that aligns slab and bulk crystal representations in a shared latent space. From a slab query, CatRetriever accurately retrieves plausible parent bulk candidates with R@1 > 91% and R@3 > 98% on both the indistribution and holdout evaluation sets. We further extend the CatRetriever framework into an adsorption energy targeted bulk discovery pipeline that combines bulk retrieval, generative search space expansion, and adsorption energy distribution analysis. This workflow evaluates candidates by both structural compatibility with the query slab and their ability to access the target adsorption energy range across diverse surface environments. CatRetriever therefore provides a scalable route for connecting catalyst generative models with physically plausible and adsorption energy compatible bulk catalyst discovery.  \n1. Introduction  \nDiscovering highly active catalysts is a central goal in enhancing the efficiency of energy conversion processes 1-3. Traditionally, these efforts have relied on experimental trialand-error and density functional theory (DFT)-based high-throughput screening, which remain time-consuming and computationally expensive, limiting the breadth of the explorable chemical space. Moreover, conventional screening is inherently constrained by predefined databases or enumerated candidate spaces, and its scalability decreases rapidly as the chemical and configurational complexity of catalyst structures increases4, 5. These challenges have motivated a shift from database-bound forward screening toward inverse design strategies. Artificial intelligence (AI) has emerged as a powerful alternative, particularly through generative models capable of proposing novel crystal structures beyond those catalogued in existing databases6-8.  \nThis generative paradigm has recently been extended to heterogeneous cata","cbCaiu1LHmH56deD","https://ap.wps.com/l/cbCaiu1LHmH56deD","pdf",2066715,3,1,45,"English","en",105,"# Introduction\n## Inverse design and generative catalyst surfaces\n## The slab-to-bulk retrieval problem\n## Motivation: bulk-dependent thermodynamic descriptors","[{\"question\":\"What gap do slab-based catalyst generative models have in catalyst discovery?\",\"answer\":\"They generate catalyst surface–adsorbate structures at the slab level but do not provide the corresponding parent bulk crystal, preventing rigorous assessment of bulk-dependent descriptors like formation energy, surface energy, and crystallographic symmetry.\"},{\"question\":\"How does CatRetriever address the slab-to-bulk retrieval problem?\",\"answer\":\"CatRetriever uses contrastive representation learning to align slab and bulk crystal representations in a shared latent space, enabling retrieval of plausible parent bulk candidates from a slab query.\"},{\"question\":\"How is CatRetriever extended for adsorption-energy targeted bulk discovery?\",\"answer\":\"The framework is combined into a pipeline that uses bulk retrieval, generative search space expansion, and adsorption energy distribution analysis to evaluate candidates by structural compatibility and by access to a target adsorption-energy range across surface environments.\"}]",1784212751,113,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"catretriever-contrastive-representation-learning-for-slab-to-bulk-retrieval-in-generative-catalyst-discovery","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/catretriever-contrastive-representation-learning-for-slab-to-bulk-retrieval-in-generative-catalyst-discovery/86579/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","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 gap do slab-based catalyst generative models have in catalyst discovery?","Question",{"text":75,"@type":76},"They generate catalyst surface–adsorbate structures at the slab level but do not provide the corresponding parent bulk crystal, preventing rigorous assessment of bulk-dependent descriptors like formation energy, surface energy, and crystallographic symmetry.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does CatRetriever address the slab-to-bulk retrieval problem?",{"text":80,"@type":76},"CatRetriever uses contrastive representation learning to align slab and bulk crystal representations in a shared latent space, enabling retrieval of plausible parent bulk candidates from a slab query.",{"name":82,"@type":73,"acceptedAnswer":83},"How is CatRetriever extended for adsorption-energy targeted bulk discovery?",{"text":84,"@type":76},"The framework is combined into a pipeline that uses bulk retrieval, generative search space expansion, and adsorption energy distribution analysis to evaluate candidates by structural compatibility and by access to a target adsorption-energy range across surface environments.","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":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]