{"id":"https://openalex.org/W7163585102","doi":"https://doi.org/10.48550/arxiv.2606.04434","title":"Hyper-ICL: Attention Calibration with Hyperbolic Anchor Distillation for Multimodal In-Context Learning","display_name":"Hyper-ICL: Attention Calibration with Hyperbolic Anchor Distillation for Multimodal In-Context Learning","publication_year":2026,"publication_date":"2026-06-03","ids":{"openalex":"https://openalex.org/W7163585102","doi":"https://doi.org/10.48550/arxiv.2606.04434"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.04434","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04434","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.04434","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5031144137","display_name":"Niloufar Alipour Talemi","orcid":"https://orcid.org/0009-0000-6881-3671"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Talemi, Niloufar Alipour","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001509911","display_name":"Hossein Kashiani","orcid":"https://orcid.org/0000-0001-8338-9987"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kashiani, Hossein","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137886283","display_name":"Fatemeh Afghah","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Afghah, Fatemeh","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":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.5421000123023987,"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":0.5421000123023987,"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":0.23010000586509705,"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/T10028","display_name":"Topic Modeling","score":0.045499999076128006,"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/inference","display_name":"Inference","score":0.6880000233650208},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4867999851703644},{"id":"https://openalex.org/keywords/margin","display_name":"Margin (machine learning)","score":0.45719999074935913},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.42170000076293945},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.41359999775886536},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.3797000050544739},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.35030001401901245},{"id":"https://openalex.org/keywords/multimodal-learning","display_name":"Multimodal learning","score":0.3497999906539917}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7204999923706055},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6880000233650208},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5019000172615051},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4867999851703644},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.45719999074935913},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.42170000076293945},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.41359999775886536},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4074000120162964},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3797000050544739},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.35030001401901245},{"id":"https://openalex.org/C2780660688","wikidata":"https://www.wikidata.org/wiki/Q25052564","display_name":"Multimodal learning","level":2,"score":0.3497999906539917},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.3472999930381775},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.3434999883174896},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.3393000066280365},{"id":"https://openalex.org/C177284502","wikidata":"https://www.wikidata.org/wiki/Q1005390","display_name":"Adapter (computing)","level":2,"score":0.32499998807907104},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2985999882221222},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.2946999967098236},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.28619998693466187},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.2768000066280365},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2624000012874603},{"id":"https://openalex.org/C115051666","wikidata":"https://www.wikidata.org/wiki/Q6522493","display_name":"Ranging","level":2,"score":0.259799987077713}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.04434","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04434","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.04434","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.04434","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"score":0.7529876828193665,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Multimodal":[0,12],"In-Context":[1,23],"Learning":[2],"(ICL)":[3],"has":[4],"emerged":[5],"as":[6],"a":[7,17,62,83,108,132,144],"practical":[8],"inference":[9,40,79],"paradigm":[10],"for":[11,66],"Large":[13],"Language":[14],"Models,":[15],"where":[16],"small":[18],"set":[19],"of":[20],"interleaved":[21],"image-text":[22],"Demonstrations":[24],"(ICDs)":[25],"conditions":[26],"the":[27,126,154,158],"model":[28],"to":[29,47,49,92,143,156],"solve":[30],"new":[31],"tasks.":[32],"Despite":[33],"its":[34],"flexibility,":[35],"multimodal":[36,68,169],"ICL":[37,69,187],"incurs":[38],"high":[39],"latency":[41],"and":[42,53,122,174,183,188],"suffers":[43],"from":[44],"instability":[45],"due":[46],"sensitivity":[48],"demonstration":[50,72,101],"formatting,":[51],"ordering,":[52],"content.":[54],"To":[55,98],"address":[56],"these":[57],"limitations,":[58],"we":[59,106,130],"propose":[60,131],"Hyper-ICL,":[61],"lightweight,":[63],"training-based":[64],"framework":[65],"demonstration-free":[67],"that":[70,88,112,138,178],"reconstructs":[71],"effects":[73],"directly":[74],"without":[75],"requiring":[76],"ICDs":[77],"at":[78,117],"time.":[80],"Hyper-ICL":[81,179],"learns":[82],"parameter-efficient":[84],"low-rank":[85],"logit-level":[86],"adapter":[87],"calibrates":[89],"attention":[90,96],"distributions":[91],"better":[93],"match":[94],"demonstration-induced":[95],"redistribution.":[97],"capture":[99],"how":[100],"influence":[102],"varies":[103],"across":[104,120,166],"queries,":[105],"introduce":[107],"query-adaptive":[109],"modulation":[110],"mechanism":[111],"adaptively":[113],"controls":[114],"intervention":[115],"strength":[116],"token":[118],"level":[119],"layers":[121],"heads":[123],"based":[124],"on":[125],"current":[127],"query.":[128],"Finally,":[129],"layer-wise":[133],"hyperbolic":[134],"anchor":[135],"distillation":[136],"loss":[137,152],"aligns":[139],"intermediate":[140],"student":[141,155],"features":[142],"demonstration-conditioned":[145],"teacher":[146],"via":[147],"Lorentz":[148],"geodesic":[149],"distance.":[150],"This":[151],"encourages":[153],"reconstruct":[157],"demonstration-query":[159],"relationships":[160],"induced":[161],"by":[162],"ICDs.":[163],"Extensive":[164],"experiments":[165],"six":[167],"different":[168],"benchmarks":[170],"(including":[171],"VQAv2,":[172],"OK-VQA,":[173],"COCO":[175],"Caption)":[176],"demonstrate":[177],"consistently":[180],"improves":[181],"accuracy":[182],"stability":[184],"over":[185],"vanilla":[186],"existing":[189],"state-of-the-art":[190],"methods.":[191]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-05T00:00:00"}