{"id":"https://openalex.org/W4415538243","doi":"https://doi.org/10.1145/3746027.3755778","title":"Evaluating and Mitigating Sycophancy in Large Vision-Language Models","display_name":"Evaluating and Mitigating Sycophancy in Large Vision-Language Models","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415538243","doi":"https://doi.org/10.1145/3746027.3755778"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3755778","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755778","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"article","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/A5106730293","display_name":"Jiayi Gao","orcid":"https://orcid.org/0009-0001-2970-4170"},"institutions":[{"id":"https://openalex.org/I2722730","display_name":"Inner Mongolia University","ror":"https://ror.org/0106qb496","country_code":"CN","type":"education","lineage":["https://openalex.org/I2722730"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Jiayi Gao","raw_affiliation_strings":["Inner Mongolia University, Hohhot, China"],"raw_orcid":"https://orcid.org/0009-0001-2970-4170","affiliations":[{"raw_affiliation_string":"Inner Mongolia University, Hohhot, China","institution_ids":["https://openalex.org/I2722730"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5000417495","display_name":"Huaiwen Zhang","orcid":"https://orcid.org/0000-0002-3183-9218"},"institutions":[{"id":"https://openalex.org/I2722730","display_name":"Inner Mongolia University","ror":"https://ror.org/0106qb496","country_code":"CN","type":"education","lineage":["https://openalex.org/I2722730"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huaiwen Zhang","raw_affiliation_strings":["Inner Mongolia University, Hohhot, China"],"raw_orcid":"https://orcid.org/0000-0002-3183-9218","affiliations":[{"raw_affiliation_string":"Inner Mongolia University, Hohhot, China","institution_ids":["https://openalex.org/I2722730"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5106730293"],"corresponding_institution_ids":["https://openalex.org/I2722730"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28744374,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11706","last_page":"11715"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9962000250816345,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9876999855041504,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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/profiling","display_name":"Profiling (computer programming)","score":0.5613999962806702},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.41909998655319214},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.3788999915122986},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.35760000348091125},{"id":"https://openalex.org/keywords/source-code","display_name":"Source code","score":0.2874999940395355}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7222999930381775},{"id":"https://openalex.org/C187191949","wikidata":"https://www.wikidata.org/wiki/Q1138496","display_name":"Profiling (computer programming)","level":2,"score":0.5613999962806702},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.4226999878883362},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.41909998655319214},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3788999915122986},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.37049999833106995},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3675000071525574},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.35760000348091125},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.30489999055862427},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.26989999413490295},{"id":"https://openalex.org/C2780589192","wikidata":"https://www.wikidata.org/wiki/Q7285140","display_name":"Raising (metalworking)","level":2,"score":0.2696000039577484},{"id":"https://openalex.org/C50335755","wikidata":"https://www.wikidata.org/wiki/Q483247","display_name":"Phenomenon","level":2,"score":0.26440000534057617},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2583000063896179}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3755778","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755778","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":3,"referenced_works":["https://openalex.org/W2277195237","https://openalex.org/W4388778348","https://openalex.org/W4409129946"],"related_works":[],"abstract_inverted_index":{"Large":[0],"vision-language":[1],"models":[2],"(LVLMs)":[3],"have":[4],"recently":[5],"achieved":[6],"significant":[7],"advancements,":[8],"demonstrating":[9],"powerful":[10],"capabilities":[11],"in":[12,72,87,96,112,142,172],"understanding":[13,98],"and":[14,48,68,99,124,156],"reasoning":[15,100],"about":[16,54],"visual":[17,97],"information.":[18],"However,":[19],"LVLMs":[20,143,173],"may":[21],"generate":[22],"biased":[23],"responses":[24],"that":[25,60,165],"reflect":[26],"the":[27,32,45,66,146,152,158,166],"user":[28],"beliefs":[29],"rather":[30],"than":[31],"facts,":[33],"a":[34,80,91,105,131],"phenomenon":[35],"known":[36],"as":[37],"sycophancy.":[38],"Sycophancy":[39],"can":[40],"pose":[41],"serious":[42],"challenges":[43],"to":[44,84,139],"performance,":[46],"trustworthiness,":[47],"security":[49],"of":[50,70,94,120,149,160],"LVLMs,":[51,115],"raising":[52],"concerns":[53],"their":[55,125],"practical":[56],"applications.":[57],"We":[58,109],"note":[59],"there":[61],"is":[62,180],"limited":[63],"work":[64],"on":[65],"evaluation":[67,93],"mitigation":[69],"sycophancy":[71,86,95,111,141,171],"LVLMs.":[73,88,161],"In":[74],"this":[75],"paper,":[76],"we":[77,129],"introduce":[78],"SyEval-VL,":[79],"benchmark":[81],"specifically":[82],"designed":[83],"evaluate":[85,110],"SyEval-VL":[89],"offers":[90],"comprehensive":[92],"across":[101],"various":[102,121],"scenarios":[103],"with":[104],"multi-round":[106],"dialogue":[107],"format.":[108],"several":[113],"popular":[114],"providing":[116],"an":[117],"in-depth":[118],"analysis":[119],"sycophantic":[122],"behaviors":[123],"consequential":[126],"impacts.":[127],"Additionally,":[128],"propose":[130],"novel":[132],"framework,":[133],"Human":[134],"Feedback-based":[135],"Retrieval-Augmented":[136],"Generation":[137],"(HFRAG),":[138],"mitigate":[140],"by":[144],"determining":[145],"appropriate":[147],"timing":[148],"retrieval,":[150],"profiling":[151],"proper":[153],"retrieval":[154],"target,":[155],"augmenting":[157],"decoding":[159],"Extensive":[162],"experiments":[163],"demonstrate":[164],"proposed":[167],"method":[168],"significantly":[169],"mitigates":[170],"without":[174],"requiring":[175],"additional":[176],"training.":[177],"Our":[178],"code":[179],"available":[181],"at:":[182],"https://github.com/immc-lab/SyEval-VL":[183]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-25T00:00:00"}