{"id":"https://openalex.org/W4413026927","doi":"https://doi.org/10.32604/cmc.2025.067750","title":"Beyond Intentions: A Critical Survey of Misalignment in LLMs","display_name":"Beyond Intentions: A Critical Survey of Misalignment in LLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4413026927","doi":"https://doi.org/10.32604/cmc.2025.067750"},"language":"en","primary_location":{"id":"doi:10.32604/cmc.2025.067750","is_oa":true,"landing_page_url":"https://doi.org/10.32604/cmc.2025.067750","pdf_url":null,"source":{"id":"https://openalex.org/S4210191605","display_name":"Computers, materials & continua/Computers, materials & continua (Print)","issn_l":"1546-2218","issn":["1546-2218","1546-2226"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers, Materials & Continua","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.32604/cmc.2025.067750","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5039828208","display_name":"Yubin Qu","orcid":"https://orcid.org/0000-0001-5222-4020"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yubin Qu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051574202","display_name":"Song Huang","orcid":"https://orcid.org/0000-0003-1385-7591"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song Huang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082772005","display_name":"Li Long","orcid":"https://orcid.org/0000-0002-1105-3963"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Long Li","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080653475","display_name":"Peng Nie","orcid":"https://orcid.org/0000-0002-5322-6324"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng Nie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5112537239","display_name":"Yao Yongming","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yongming Yao","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5039828208"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":11.7121,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.98274361,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":95,"max":99},"biblio":{"volume":"85","issue":"1","first_page":"249","last_page":"300"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12478","display_name":"Wikis in Education and Collaboration","score":0.7741000056266785,"subfield":{"id":"https://openalex.org/subfields/3315","display_name":"Communication"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12478","display_name":"Wikis in Education and Collaboration","score":0.7741000056266785,"subfield":{"id":"https://openalex.org/subfields/3315","display_name":"Communication"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/psychology","display_name":"Psychology","score":0.4479680061340332},{"id":"https://openalex.org/keywords/survey-research","display_name":"Survey research","score":0.4113101363182068},{"id":"https://openalex.org/keywords/social-psychology","display_name":"Social psychology","score":0.3925703763961792},{"id":"https://openalex.org/keywords/political-science","display_name":"Political science","score":0.3351491689682007},{"id":"https://openalex.org/keywords/applied-psychology","display_name":"Applied psychology","score":0.22476783394813538}],"concepts":[{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.4479680061340332},{"id":"https://openalex.org/C173481278","wikidata":"https://www.wikidata.org/wiki/Q7257997","display_name":"Survey research","level":2,"score":0.4113101363182068},{"id":"https://openalex.org/C77805123","wikidata":"https://www.wikidata.org/wiki/Q161272","display_name":"Social psychology","level":1,"score":0.3925703763961792},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.3351491689682007},{"id":"https://openalex.org/C75630572","wikidata":"https://www.wikidata.org/wiki/Q538904","display_name":"Applied psychology","level":1,"score":0.22476783394813538}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.32604/cmc.2025.067750","is_oa":true,"landing_page_url":"https://doi.org/10.32604/cmc.2025.067750","pdf_url":null,"source":{"id":"https://openalex.org/S4210191605","display_name":"Computers, materials & continua/Computers, materials & continua (Print)","issn_l":"1546-2218","issn":["1546-2218","1546-2226"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers, Materials & Continua","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.32604/cmc.2025.067750","is_oa":true,"landing_page_url":"https://doi.org/10.32604/cmc.2025.067750","pdf_url":null,"source":{"id":"https://openalex.org/S4210191605","display_name":"Computers, materials & continua/Computers, materials & continua (Print)","issn_l":"1546-2218","issn":["1546-2218","1546-2226"],"is_oa":true,"is_in_doaj":false,"is_core":true,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computers, Materials & Continua","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/10","display_name":"Reduced inequalities","score":0.6600000262260437}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":27,"referenced_works":["https://openalex.org/W2979417040","https://openalex.org/W3002093512","https://openalex.org/W3201174429","https://openalex.org/W4285206226","https://openalex.org/W4293458267","https://openalex.org/W4385452929","https://openalex.org/W4385572928","https://openalex.org/W4389519291","https://openalex.org/W4392353733","https://openalex.org/W4399750638","https://openalex.org/W4401043050","https://openalex.org/W4401389494","https://openalex.org/W4401450595","https://openalex.org/W4402455807","https://openalex.org/W4402670459","https://openalex.org/W4402671221","https://openalex.org/W4406222181","https://openalex.org/W4406882309","https://openalex.org/W4407159610","https://openalex.org/W4408253143","https://openalex.org/W4411120032","https://openalex.org/W4411120457","https://openalex.org/W4411337523","https://openalex.org/W4411551657","https://openalex.org/W4412888492","https://openalex.org/W4415796863","https://openalex.org/W4415797408"],"related_works":["https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W4391375266","https://openalex.org/W2931662336","https://openalex.org/W4220667126","https://openalex.org/W2077865380","https://openalex.org/W3006817050","https://openalex.org/W4401768695","https://openalex.org/W2765597752","https://openalex.org/W2134894512"],"abstract_inverted_index":{"Large":[0],"language":[1],"models":[2],"(LLMs)":[3],"represent":[4],"significant":[5],"advancements":[6],"in":[7,183],"artificial":[8],"intelligence.":[9],"However,":[10],"their":[11],"increasing":[12],"capabilities":[13],"come":[14],"with":[15,53,275],"a":[16,54],"serious":[17],"challenge:":[18],"misalignment,":[19,94,139,155],"which":[20,199],"refers":[21],"to":[22,38,79,110,153,271],"the":[23,29,40,44,59,90,136,201,210,250,259],"deviation":[24],"of":[25,43,61,93,138,203,254,263],"model":[26,108,197,204],"behavior":[27],"from":[28,98],"designers\u2019":[30],"intentions":[31],"and":[32,48,51,63,85,122,129,134,145,169,175,196,208,218,228,247,261],"human":[33,276],"values.":[34,277],"This":[35,125,256],"review":[36,126],"aims":[37],"synthesize":[39],"current":[41,219],"understanding":[42],"LLM":[45,264],"misalignment":[46,62,184,265],"issue":[47],"provide":[49],"researchers":[50],"practitioners":[52],"comprehensive":[55],"overview.":[56],"We":[57,88,206],"define":[58],"concept":[60],"elaborate":[64],"on":[65,157],"its":[66],"various":[67],"manifestations,":[68],"including":[69,187,235],"generating":[70],"harmful":[71],"content,":[72],"factual":[73],"errors":[74],"(hallucinations),":[75],"propagating":[76],"biases,":[77],"failing":[78],"follow":[80],"instructions,":[81],"emerging":[82,130],"deceptive":[83],"behaviors,":[84],"emergent":[86],"misalignment.":[87],"explore":[89],"multifaceted":[91],"causes":[92],"systematically":[95],"analyzing":[96],"factors":[97],"surface-level":[99],"technical":[100],"issues":[101],"(e.g.,":[102,114],"training":[103,120],"data,":[104],"objective":[105],"function":[106],"design,":[107],"scaling)":[109],"deeper":[111],"fundamental":[112],"challenges":[113],"difficulties":[115],"formalizing":[116],"values,":[117,246],"discrepancies":[118],"between":[119],"signals":[121],"real":[123],"intentions).":[124],"covers":[127],"existing":[128],"techniques":[131,160],"for":[132,268],"detecting":[133],"evaluating":[135],"degree":[137],"such":[140,161],"as":[141,162],"benchmark":[142],"tests,":[143],"red-teaming,":[144],"formal":[146],"safety":[147],"assessments.":[148],"Subsequently,":[149],"we":[150,179,223],"examine":[151],"strategies":[152],"mitigate":[154],"focusing":[156],"mainstream":[158],"alignment":[159,238,242],"RLHF,":[163],"Constitutional":[164],"AI":[165],"(CAI),":[166],"instruction":[167],"fine-tuning,":[168,192],"novel":[170,241],"approaches":[171,270],"that":[172],"address":[173],"scalability":[174],"robustness.":[176],"In":[177],"particular,":[178],"analyze":[180,209],"recent":[181],"advances":[182],"attack":[185],"research,":[186],"system":[188],"prompt":[189],"modifications,":[190],"supervised":[191],"self-supervised":[193],"representation":[194],"attacks,":[195],"editing,":[198],"challenge":[200],"robustness":[202],"alignment.":[205,255],"categorize":[207],"surveyed":[211],"literature,":[212],"highlighting":[213],"major":[214],"findings,":[215],"persistent":[216],"limitations,":[217],"contentious":[220],"points.":[221],"Finally,":[222],"identify":[224],"key":[225],"open":[226],"questions":[227],"propose":[229],"several":[230],"promising":[231],"future":[232],"research":[233],"directions,":[234],"constructing":[236],"high-quality":[237],"datasets,":[239],"exploring":[240],"methods,":[243],"coordinating":[244],"diverse":[245],"delving":[248],"into":[249],"deep":[251],"philosophical":[252],"aspects":[253],"work":[257],"underscores":[258],"complexity":[260],"multidimensionality":[262],"issues,":[266],"calling":[267],"interdisciplinary":[269],"reliably":[272],"align":[273],"LLMs":[274]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":2}],"updated_date":"2026-03-27T14:29:43.386196","created_date":"2025-10-10T00:00:00"}