{"id":"https://openalex.org/W4412889574","doi":"https://doi.org/10.18653/v1/2025.acl-long.1542","title":"Efficient Many-Shot In-Context Learning with Dynamic Block-Sparse Attention","display_name":"Efficient Many-Shot In-Context Learning with Dynamic Block-Sparse Attention","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889574","doi":"https://doi.org/10.18653/v1/2025.acl-long.1542"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-long.1542","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1542","pdf_url":"https://aclanthology.org/2025.acl-long.1542.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-long.1542.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102777466","display_name":"Emily Xiao","orcid":"https://orcid.org/0000-0001-8895-771X"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Emily Xiao","raw_affiliation_strings":["Language Technologies Institute , Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Language Technologies Institute , Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023430268","display_name":"C. H. Li","orcid":null},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chin-Jou Li","raw_affiliation_strings":["Language Technologies Institute , Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Language Technologies Institute , Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100758237","display_name":"Yilin Zhang","orcid":"https://orcid.org/0000-0002-8238-2326"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yilin Zhang","raw_affiliation_strings":["Language Technologies Institute , Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Language Technologies Institute , Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068811427","display_name":"Graham Neubig","orcid":"https://orcid.org/0000-0002-2072-3789"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Graham Neubig","raw_affiliation_strings":["Language Technologies Institute , Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Language Technologies Institute , Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5013007353","display_name":"Amanda Bertsch","orcid":"https://orcid.org/0000-0002-1368-1111"},"institutions":[{"id":"https://openalex.org/I74973139","display_name":"Carnegie Mellon University","ror":"https://ror.org/05x2bcf33","country_code":"US","type":"education","lineage":["https://openalex.org/I74973139"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Amanda Bertsch","raw_affiliation_strings":["Language Technologies Institute , Carnegie Mellon University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Language Technologies Institute , Carnegie Mellon University","institution_ids":["https://openalex.org/I74973139"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I74973139"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"31946","last_page":"31958"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9794999957084656,"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"}},"topics":[{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9794999957084656,"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/computer-science","display_name":"Computer science","score":0.7151778936386108},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5802888870239258},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.5701380372047424},{"id":"https://openalex.org/keywords/shot","display_name":"Shot (pellet)","score":0.5590603351593018},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.38708651065826416},{"id":"https://openalex.org/keywords/human\u2013computer-interaction","display_name":"Human\u2013computer interaction","score":0.32564765214920044},{"id":"https://openalex.org/keywords/materials-science","display_name":"Materials science","score":0.08908677101135254},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.07888719439506531},{"id":"https://openalex.org/keywords/geology","display_name":"Geology","score":0.06390640139579773}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7151778936386108},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5802888870239258},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.5701380372047424},{"id":"https://openalex.org/C2778344882","wikidata":"https://www.wikidata.org/wiki/Q278938","display_name":"Shot (pellet)","level":2,"score":0.5590603351593018},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.38708651065826416},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.32564765214920044},{"id":"https://openalex.org/C192562407","wikidata":"https://www.wikidata.org/wiki/Q228736","display_name":"Materials science","level":0,"score":0.08908677101135254},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.07888719439506531},{"id":"https://openalex.org/C127313418","wikidata":"https://www.wikidata.org/wiki/Q1069","display_name":"Geology","level":0,"score":0.06390640139579773},{"id":"https://openalex.org/C191897082","wikidata":"https://www.wikidata.org/wiki/Q11467","display_name":"Metallurgy","level":1,"score":0.0},{"id":"https://openalex.org/C2524010","wikidata":"https://www.wikidata.org/wiki/Q8087","display_name":"Geometry","level":1,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.acl-long.1542","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1542","pdf_url":"https://aclanthology.org/2025.acl-long.1542.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-long.1542","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-long.1542","pdf_url":"https://aclanthology.org/2025.acl-long.1542.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":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1778200954","display_name":null,"funder_award_id":"DGE2140739","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889574.pdf","grobid_xml":"https://content.openalex.org/works/W4412889574.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W2074502265","https://openalex.org/W4214877189","https://openalex.org/W2773965352","https://openalex.org/W2381179799","https://openalex.org/W4411535265","https://openalex.org/W2980279061","https://openalex.org/W2334685461","https://openalex.org/W2366718574","https://openalex.org/W2359774528","https://openalex.org/W4298312966"],"abstract_inverted_index":{"Many-shot":[0],"in-context":[1,69],"learning":[2],"has":[3],"recently":[4],"shown":[5],"promise":[6],"as":[7],"an":[8],"alternative":[9],"to":[10,33,41,88],"finetuning,":[11],"with":[12],"the":[13,17,28,96,112],"major":[14],"advantage":[15],"that":[16,107],"same":[18],"model":[19],"can":[20],"be":[21],"served":[22],"for":[23,55,66],"multiple":[24],"tasks.However,":[25],"this":[26,108],"shifts":[27],"computational":[29],"burden":[30],"from":[31],"training-time":[32],"inference-time,":[34],"making":[35],"deployment":[36,113],"of":[37,78,81,95,114],"many-shot":[38,68,115],"ICL":[39,102,116],"challenging":[40],"justify":[42],"in-practice.This":[43],"cost":[44],"is":[45,53],"further":[46,110],"increased":[47],"if":[48],"a":[49,63],"custom":[50],"demonstration":[51],"set":[52],"retrieved":[54],"each":[56],"inference":[57],"example.We":[58],"present":[59],"Dynamic":[60],"Block-Sparse":[61],"Attention,":[62],"training-free":[64],"framework":[65],"retrieval-based":[67],"learning.By":[70],"combining":[71],"carefully":[72],"designed":[73],"blocksparse":[74],"attention":[75],"and":[76,103],"retrieval":[77],"cached":[79],"groups":[80],"demonstrations,":[82],"we":[83],"achieve":[84],"comparable":[85],"perexample":[86],"latency":[87],"finetuning":[89,104],"while":[90],"maintaining":[91],"on":[92],"average":[93],">95%":[94],"best":[97],"method's":[98],"accuracy":[99],"across":[100],"strong":[101],"baselines.We":[105],"hope":[106],"will":[109],"enable":[111],"at":[117],"scale.":[118],"1":[119]},"counts_by_year":[{"year":2025,"cited_by_count":1}],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}