{"id":"https://openalex.org/W7164801340","doi":"https://doi.org/10.1145/3805622.3810627","title":"DyCa-GRPO:Calibrated and Efficient Learning-to-Rank for Multimodal Retrieval with Human Feedback","display_name":"DyCa-GRPO:Calibrated and Efficient Learning-to-Rank for Multimodal Retrieval with Human Feedback","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164801340","doi":"https://doi.org/10.1145/3805622.3810627"},"language":null,"primary_location":{"id":"doi:10.1145/3805622.3810627","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810627","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3805622.3810627","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138677123","display_name":"Ning Han","orcid":"https://orcid.org/0009-0006-4277-6511"},"institutions":[{"id":"https://openalex.org/I109935558","display_name":"Ningbo University","ror":"https://ror.org/03et85d35","country_code":"CN","type":"education","lineage":["https://openalex.org/I109935558"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ning Han","raw_affiliation_strings":["School of Software Technology, Zhejiang University, Ningbo, China"],"raw_orcid":"https://orcid.org/0009-0006-4277-6511","affiliations":[{"raw_affiliation_string":"School of Software Technology, Zhejiang University, Ningbo, China","institution_ids":["https://openalex.org/I109935558","https://openalex.org/I76130692"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043803804","display_name":"Xiubo Liang","orcid":"https://orcid.org/0000-0002-4749-5552"},"institutions":[{"id":"https://openalex.org/I109935558","display_name":"Ningbo University","ror":"https://ror.org/03et85d35","country_code":"CN","type":"education","lineage":["https://openalex.org/I109935558"]},{"id":"https://openalex.org/I76130692","display_name":"Zhejiang University","ror":"https://ror.org/00a2xv884","country_code":"CN","type":"education","lineage":["https://openalex.org/I76130692"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiubo Liang","raw_affiliation_strings":["School of Software Technology, Zhejiang University, Ningbo, China"],"raw_orcid":"https://orcid.org/0000-0002-4749-5552","affiliations":[{"raw_affiliation_string":"School of Software Technology, Zhejiang University, Ningbo, China","institution_ids":["https://openalex.org/I109935558","https://openalex.org/I76130692"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.9330206,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"215","last_page":"222"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.7179999947547913,"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":0.7179999947547913,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.09549999982118607,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10028","display_name":"Topic Modeling","score":0.03869999945163727,"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/robustness","display_name":"Robustness (evolution)","score":0.6402000188827515},{"id":"https://openalex.org/keywords/learning-to-rank","display_name":"Learning to rank","score":0.5733000040054321},{"id":"https://openalex.org/keywords/relevance","display_name":"Relevance (law)","score":0.5626000165939331},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.5322999954223633},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.5239999890327454},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.44670000672340393},{"id":"https://openalex.org/keywords/preference","display_name":"Preference","score":0.429500013589859},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.4043999910354614}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.809499979019165},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6402000188827515},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6018000245094299},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6004999876022339},{"id":"https://openalex.org/C86037889","wikidata":"https://www.wikidata.org/wiki/Q4330127","display_name":"Learning to rank","level":3,"score":0.5733000040054321},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.5626000165939331},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.5322999954223633},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.5239999890327454},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.44670000672340393},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.429500013589859},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.4043999910354614},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.39010000228881836},{"id":"https://openalex.org/C2776321320","wikidata":"https://www.wikidata.org/wiki/Q857525","display_name":"Annotation","level":2,"score":0.38989999890327454},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.37290000915527344},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.3702000081539154},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3675000071525574},{"id":"https://openalex.org/C2779532271","wikidata":"https://www.wikidata.org/wiki/Q445558","display_name":"Relevance feedback","level":4,"score":0.35839998722076416},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.32519999146461487},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.31279999017715454},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2768000066280365},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.274399995803833},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.26489999890327454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3805622.3810627","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810627","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3805622.3810627","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3805622.3810627","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 2026 International Conference on Multimedia Retrieval","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":8,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2096733369","https://openalex.org/W2138621090","https://openalex.org/W2254249950","https://openalex.org/W3099700870","https://openalex.org/W4387848573","https://openalex.org/W4403713311","https://openalex.org/W4414281281"],"related_works":[],"abstract_inverted_index":{"Learning":[0],"to":[1,10,32,103],"rank":[2],"multimodal":[3,38,67,101,153],"content\u2014such":[4],"as":[5],"images,":[6],"videos,":[7],"and":[8,48,90,94,108,116,126,139,151],"text\u2014according":[9],"human":[11,39],"preferences":[12],"is":[13],"a":[14,100,156],"fundamental":[15],"challenge":[16],"in":[17,37,130],"multimedia":[18],"retrieval.":[19,68],"Existing":[20],"preference":[21,149],"optimization":[22,150],"methods,":[23],"originally":[24],"designed":[25],"for":[26,66,159],"text-only":[27],"language":[28],"models,":[29],"often":[30],"fail":[31],"handle":[33],"the":[34],"inherent":[35],"noise":[36,89],"feedback,":[40],"suffer":[41],"from":[42],"poor":[43],"calibration":[44,134],"of":[45],"relevance":[46,85,110],"scores,":[47],"inefficiently":[49],"utilize":[50],"limited":[51],"annotation":[52],"budgets.":[53],"To":[54],"address":[55],"these":[56],"issues,":[57],"we":[58],"propose":[59],"DyCa-GRPO":[60,122],"(Dynamic":[61],"Calibration\u2013enhanced":[62],"Group-wise":[63],"Preference":[64],"Optimization)":[65],"Our":[69,146],"approach":[70],"introduces":[71],"(1)":[72],"dynamic":[73],"sampling":[74],"with":[75,118],"margin":[76],"filtering,":[77],"which":[78,98],"constructs":[79],"training":[80],"pairs":[81],"only":[82],"when":[83],"human-rated":[84],"scores":[86],"differ":[87],"significantly\u2014reducing":[88],"improving":[91],"sample":[92],"efficiency;":[93],"(2)":[95],"score-calibrated":[96],"learning,":[97],"trains":[99],"ranker":[102],"jointly":[104],"optimize":[105],"relative":[106],"ranking":[107],"absolute":[109],"score":[111],"prediction.":[112],"Evaluated":[113],"on":[114],"COCO":[115],"WebVision":[117],"human-derived":[119],"quality":[120],"annotations,":[121],"outperforms":[123],"DPO,":[124],"PPO,":[125],"GRPO":[127],"by":[128,137],"4.3\u20136.8%":[129],"nDCG@10,":[131],"reduces":[132],"expected":[133],"error":[135],"(ECE)":[136],"35%,":[138],"maintains":[140],"robustness":[141],"under":[142],"20%":[143],"label":[144],"noise.":[145],"work":[147],"bridges":[148],"calibrated":[152],"retrieval,":[154],"offering":[155],"practical":[157],"framework":[158],"human-aligned":[160],"search":[161],"systems.":[162]},"counts_by_year":[],"updated_date":"2026-06-16T07:37:23.134862","created_date":"2026-06-16T00:00:00"}