{"id":"https://openalex.org/W4416034846","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.469","title":"Enhancing LLM Knowledge Learning through Generalization","display_name":"Enhancing LLM Knowledge Learning through Generalization","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416034846","doi":"https://doi.org/10.18653/v1/2025.findings-emnlp.469"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-emnlp.469","is_oa":false,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.469","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112897851","display_name":"Mingkang Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingkang Zhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5081740800","display_name":"Xi Chen","orcid":"https://orcid.org/0000-0002-8057-4623"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xi Chen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063901724","display_name":"Zhongdao Wang","orcid":"https://orcid.org/0000-0002-4483-8783"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhongdao Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051340429","display_name":"Bei Yu","orcid":"https://orcid.org/0000-0001-6406-4810"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bei Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113288135","display_name":"Hengshuang Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hengshuang Zhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5052856441","display_name":"Jiaya Jia","orcid":"https://orcid.org/0000-0002-1246-553X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaya Jia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"8842","last_page":"8855"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5662999749183655,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.5662999749183655,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T13629","display_name":"Text Readability and Simplification","score":0.1665000021457672,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.08699999749660492,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.7135000228881836},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.4163999855518341},{"id":"https://openalex.org/keywords/empirical-research","display_name":"Empirical research","score":0.3560999929904938},{"id":"https://openalex.org/keywords/domain-knowledge","display_name":"Domain knowledge","score":0.3513999879360199},{"id":"https://openalex.org/keywords/knowledge-based-systems","display_name":"Knowledge-based systems","score":0.33880001306533813},{"id":"https://openalex.org/keywords/knowledge-extraction","display_name":"Knowledge extraction","score":0.33219999074935913},{"id":"https://openalex.org/keywords/empirical-evidence","display_name":"Empirical evidence","score":0.32589998841285706},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.3231000006198883},{"id":"https://openalex.org/keywords/procedural-knowledge","display_name":"Procedural knowledge","score":0.3118000030517578}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7675999999046326},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.7135000228881836},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4848000109195709},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.4163999855518341},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.3560999929904938},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.3513999879360199},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3418999910354614},{"id":"https://openalex.org/C115925183","wikidata":"https://www.wikidata.org/wiki/Q1412694","display_name":"Knowledge-based systems","level":2,"score":0.33880001306533813},{"id":"https://openalex.org/C120567893","wikidata":"https://www.wikidata.org/wiki/Q1582085","display_name":"Knowledge extraction","level":2,"score":0.33219999074935913},{"id":"https://openalex.org/C166052673","wikidata":"https://www.wikidata.org/wiki/Q83021","display_name":"Empirical evidence","level":2,"score":0.32589998841285706},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.3231000006198883},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.3174999952316284},{"id":"https://openalex.org/C124469403","wikidata":"https://www.wikidata.org/wiki/Q1813993","display_name":"Procedural knowledge","level":3,"score":0.3118000030517578},{"id":"https://openalex.org/C2776291881","wikidata":"https://www.wikidata.org/wiki/Q6423378","display_name":"Knowledge level","level":2,"score":0.3050999939441681},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.2750999927520752},{"id":"https://openalex.org/C2777220311","wikidata":"https://www.wikidata.org/wiki/Q6423340","display_name":"Knowledge acquisition","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C56814567","wikidata":"https://www.wikidata.org/wiki/Q1323686","display_name":"Explicit knowledge","level":2,"score":0.2712000012397766},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C2986065213","wikidata":"https://www.wikidata.org/wiki/Q743861","display_name":"Implicit knowledge","level":2,"score":0.2703000009059906},{"id":"https://openalex.org/C84685590","wikidata":"https://www.wikidata.org/wiki/Q1540472","display_name":"Knowledge engineering","level":2,"score":0.26910001039505005},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2689000070095062},{"id":"https://openalex.org/C56289545","wikidata":"https://www.wikidata.org/wiki/Q6423376","display_name":"Knowledge integration","level":3,"score":0.26669999957084656},{"id":"https://openalex.org/C14156362","wikidata":"https://www.wikidata.org/wiki/Q3235388","display_name":"Descriptive knowledge","level":2,"score":0.26170000433921814},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C186528591","wikidata":"https://www.wikidata.org/wiki/Q619544","display_name":"Body of knowledge","level":2,"score":0.25870001316070557},{"id":"https://openalex.org/C2777348039","wikidata":"https://www.wikidata.org/wiki/Q6423397","display_name":"Knowledge space","level":2,"score":0.2513999938964844}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.findings-emnlp.469","is_oa":false,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-emnlp.469","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: EMNLP 2025","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-168965","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-168965","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Conference paper"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"As":[0],"Large":[1],"language":[2],"models":[3,18,49],"(LLMs)":[4],"are":[5,196],"increasingly":[6],"deployed":[7],"in":[8,175],"diverse":[9,83,108],"applications,":[10],"faithfully":[11],"integrating":[12],"evolving":[13],"factual":[14,60,79,155],"knowledge":[15,34,80,95,123,130,144],"into":[16],"these":[17],"remains":[19],"a":[20],"critical":[21],"challenge.":[22],"Continued":[23],"pre-training":[24,178],"on":[25,47,99],"paraphrased":[26,84,109,191],"data":[27,136,195],"has":[28],"shown":[29],"empirical":[30],"promise":[31],"for":[32,53],"enhancing":[33,129],"acquisition.":[35,131],"However,":[36],"this":[37,63,100],"approach":[38],"is":[39,86],"often":[40],"costly":[41],"and":[42,55,67,102,179,182,194],"unreliable,":[43],"as":[44,162],"it":[45],"relies":[46],"external":[48],"or":[50],"manual":[51],"effort":[52],"rewriting,":[54],"may":[56],"inadvertently":[57],"alter":[58],"the":[59,77,121,142,163],"content.":[61],"In":[62],"work,":[64],"we":[65,111,133,158],"hypothesize":[66],"empirically":[68],"show":[69],"that":[70,94],"an":[71],"LLM\u2019s":[72],"ability":[73,118],"to":[74,92,104,107,115,119,165],"continually":[75],"predict":[76,120],"same":[78,122,143],"tokens":[81,124],"given":[82,125],"contexts":[85],"positively":[87],"correlated":[88],"with":[89,190],"its":[90],"capacity":[91],"extract":[93],"via":[96],"question-answering.":[97],"Based":[98],"view":[101],"aiming":[103],"improve":[105,167],"generalization":[106],"contexts,":[110,127],"introduce":[112],"two":[113],"strategies":[114],"enhance":[116],"LLMs\u2019":[117],"varied":[126],"thereby":[128,153],"First,":[132],"propose":[134],"formatting-based":[135],"augmentation,":[137],"which":[138],"diversifies":[139],"documents":[140],"conveying":[141],"by":[145,188],"altering":[146],"document":[147],"formats":[148],"rather":[149],"than":[150],"their":[151],"content,":[152],"preserving":[154],"integrity.":[156],"Second,":[157],"adopt":[159],"sharpness-aware":[160],"minimization":[161],"optimizer":[164],"better":[166],"generalization.":[168],"Extensive":[169],"experiments":[170],"demonstrate":[171],"our":[172],"methods\u2019":[173],"effectiveness":[174],"both":[176],"continued":[177],"instruction":[180],"tuning,":[181],"further":[183],"gains":[184],"can":[185],"be":[186],"achieved":[187],"combining":[189],"data.":[192],"Code":[193],"available":[197],"at":[198],"https://github.com/dvlab-research/":[199],"llm-knowledge-generalization.":[200]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-11-08T00:00:00"}