{"id":"https://openalex.org/W7166881271","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1759","title":"SycoBench-600: Measuring Sycophancy and Correction Selectivity in LLM Assistants","display_name":"SycoBench-600: Measuring Sycophancy and Correction Selectivity in LLM Assistants","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7166881271","doi":"https://doi.org/10.18653/v1/2026.findings-acl.1759"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2026.findings-acl.1759","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1759","pdf_url":"https://aclanthology.org/2026.findings-acl.1759.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2026.findings-acl.1759.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5038099542","display_name":"Debu Sinha","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Debu Sinha","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5038099542"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.79653487,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"35278","last_page":"35284"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.07609999924898148,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.07609999924898148,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T12127","display_name":"Software System Performance and Reliability","score":0.0681999996304512,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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":0.05900000035762787,"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/selectivity","display_name":"Selectivity","score":0.390500009059906},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.3021000027656555},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.24120000004768372},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.22589999437332153}],"concepts":[{"id":"https://openalex.org/C118792377","wikidata":"https://www.wikidata.org/wiki/Q108584245","display_name":"Selectivity","level":3,"score":0.390500009059906},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.3765999972820282},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3021000027656555},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.30059999227523804},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.28700000047683716},{"id":"https://openalex.org/C136229726","wikidata":"https://www.wikidata.org/wiki/Q327092","display_name":"Biomedical engineering","level":1,"score":0.2800000011920929},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.24120000004768372},{"id":"https://openalex.org/C71924100","wikidata":"https://www.wikidata.org/wiki/Q11190","display_name":"Medicine","level":0,"score":0.22910000383853912},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.22589999437332153},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2248000055551529}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2026.findings-acl.1759","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1759","pdf_url":"https://aclanthology.org/2026.findings-acl.1759.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2026.findings-acl.1759","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2026.findings-acl.1759","pdf_url":"https://aclanthology.org/2026.findings-acl.1759.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2026","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7166881271.pdf","grobid_xml":"https://content.openalex.org/works/W7166881271.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Modern":[0],"instruction-following":[1],"language":[2],"models":[3,33],"are":[4],"optimized":[5],"to":[6,63,80,144],"be":[7],"helpful":[8],"and":[9,18,50,69,74,103,107,127,136,139,157,163],"cooperative,":[10],"often":[11],"through":[12],"preference-based":[13],"alignment":[14],"such":[15],"as":[16],"RLHF":[17],"related":[19],"methods.A":[20],"growing":[21],"body":[22],"of":[23,116],"evidence":[24],"shows":[25],"that":[26,59,142,159],"this":[27],"training":[28],"can":[29],"also":[30],"induce":[31],"sycophancy:":[32],"may":[34],"agree":[35],"with":[36],"a":[37,55],"user":[38,42],"even":[39],"when":[40],"the":[41,78,117,166],"is":[43],"wrong,":[44],"undermining":[45],"reliability":[46],"in":[47,133],"decision":[48],"support":[49],"high-stakes":[51],"advice.We":[52],"introduce":[53],"SycoBench-600,":[54],"controlled":[56],"multiple-choice":[57],"benchmark":[58,89],"measures":[60],"(i)":[61],"susceptibility":[62],"three":[64],"socialpressure":[65],"perturbations":[66],"(doubt,":[67],"authority,":[68],"an":[70],"explicit":[71],"wrong":[72],"suggestion)":[73],"(ii)":[75],"correction":[76],"selectivity,":[77],"ability":[79],"accept":[81],"correct":[82],"suggestions":[83],"while":[84],"resisting":[85],"incorrect":[86],"ones.The":[87],"released":[88],"contains":[90],"600":[91],"English":[92],"MCQ":[93],"instances":[94],"over":[95],"272":[96],"normalized":[97],"question":[98],"stems,":[99],"covers":[100],"8":[101],"domains":[102],"3":[104,112],"difficulty":[105],"tiers,":[106],"evaluates":[108],"each":[109],"instance":[110],"under":[111],"fixed":[113],"paraphrase":[114],"variants":[115],"perturbation":[118],"prompts.We":[119],"evaluate":[120],"seven":[121],"widely":[122],"used":[123],"assistants":[124],"spanning":[125],"proprietary":[126],"open-weight":[128],"families.Results":[129],"show":[130,141],"substantial":[131],"variation":[132],"pressure":[134],"robustness":[135],"selective":[137],"updating,":[138],"further":[140],"willingness":[143],"update":[145],"does":[146],"not":[147],"by":[148],"itself":[149],"imply":[150],"selectivity.We":[151],"release":[152],"raw":[153],"logs,":[154],"validation":[155],"scripts,":[156],"code":[158],"regenerates":[160],"every":[161],"table":[162],"figure":[164],"from":[165],"model":[167],"outputs.":[168]},"counts_by_year":[],"updated_date":"2026-08-22T07:34:49.880490","created_date":"2026-07-02T00:00:00"}