{"id":"https://openalex.org/W4416035557","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1432","title":"Scaling Up Temporal Domain Generalization via Temporal Experts Averaging","display_name":"Scaling Up Temporal Domain Generalization via Temporal Experts Averaging","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4416035557","doi":"https://doi.org/10.18653/v1/2025.emnlp-main.1432"},"language":null,"primary_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1432","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1432","pdf_url":"https://aclanthology.org/2025.emnlp-main.1432.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.emnlp-main.1432.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5067467814","display_name":"Aoming Liu","orcid":"https://orcid.org/0009-0007-2990-9671"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Aoming Liu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066227950","display_name":"Kevin Miller","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kevin Miller","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048704387","display_name":"Venkatesh Saligrama","orcid":"https://orcid.org/0000-0002-0675-2268"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Venkatesh Saligrama","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075906727","display_name":"Kate Saenko","orcid":"https://orcid.org/0000-0002-7564-7218"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kate Saenko","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017319429","display_name":"Boqing Gong","orcid":"https://orcid.org/0000-0003-3915-5977"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Boqing Gong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113969216","display_name":"Ser-Nam Lim","orcid":"https://orcid.org/0000-0001-9525-0288"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ser-Nam Lim","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5061227594","display_name":"Bryan A. Plummer","orcid":"https://orcid.org/0000-0002-7074-3219"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bryan A. Plummer","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5067467814"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.35100919,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"28196","last_page":"28219"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.16660000383853912,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.16660000383853912,"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/T11596","display_name":"Constraint Satisfaction and Optimization","score":0.10540000349283218,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.08240000158548355,"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/generalization","display_name":"Generalization","score":0.58160001039505},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.5034999847412109},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.47609999775886536},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3531999886035919},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.3181999921798706},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.30300000309944153}],"concepts":[{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.58160001039505},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5443000197410583},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.5034999847412109},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4860999882221222},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.47609999775886536},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40880000591278076},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3531999886035919},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.33970001339912415},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.3181999921798706},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.30300000309944153},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.28540000319480896},{"id":"https://openalex.org/C103824480","wikidata":"https://www.wikidata.org/wiki/Q185889","display_name":"Time domain","level":2,"score":0.2687000036239624},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.2540999948978424},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.2533000111579895}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.emnlp-main.1432","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1432","pdf_url":"https://aclanthology.org/2025.emnlp-main.1432.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.emnlp-main.1432","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.emnlp-main.1432","pdf_url":"https://aclanthology.org/2025.emnlp-main.1432.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 2025 Conference on Empirical Methods in Natural Language Processing","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4416035557.pdf","grobid_xml":"https://content.openalex.org/works/W4416035557.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Temporal":[0,56],"Domain":[1],"Generalization":[2],"(TDG)":[3],"aims":[4],"to":[5,47,74,86,92,146,168,173],"generalize":[6],"across":[7,150],"temporal":[8,113,131],"distribution":[9],"shifts,":[10],"e.g.,":[11],"lexical":[12],"change":[13],"over":[14],"time.Prior":[15],"work":[16],"often":[17],"addresses":[18],"this":[19],"by":[20,45,105,166],"predicting":[21],"future":[22,93,147],"model":[23,26,50,70,110],"weights.However,":[24],"full":[25],"prediction":[27],"is":[28],"prohibitively":[29],"expensive":[30],"for":[31],"even":[32],"reasonably":[33],"sized":[34],"models.Thus,":[35],"recent":[36],"methods":[37,165],"only":[38],"predict":[39],"the":[40,68,121],"classifier":[41],"layer,":[42],"limiting":[43],"generalization":[44,76,91],"failing":[46],"adjust":[48],"other":[49],"components.To":[51],"address":[52],"this,":[53],"we":[54,95,119],"propose":[55],"Experts":[57],"Averaging":[58],"(TEA),":[59],"a":[60,107,135],"novel":[61],"and":[62,156],"scalable":[63],"TDG":[64,152,158,164],"framework":[65],"that":[66,89],"updates":[67],"entire":[69],"using":[71],"weight":[72,117,132],"averaging":[73,126],"maximize":[75],"potential":[77],"while":[78,115,170],"minimizing":[79],"computational":[80],"costs.Our":[81],"theoretical":[82],"analysis":[83],"guides":[84],"us":[85],"two":[87],"steps":[88],"enhance":[90],"domains.First,":[94],"create":[96],"expert":[97],"models":[98],"with":[99],"functional":[100],"diversity":[101],"yet":[102],"parameter":[103],"similarity":[104],"fine-tuning":[106],"domain-agnostic":[108],"base":[109],"on":[111,142],"individual":[112],"domains":[114],"constraining":[116],"changes.Second,":[118],"optimize":[120],"bias-variance":[122],"tradeoff":[123],"through":[124],"adaptive":[125],"coefficients":[127],"derived":[128],"from":[129],"modeling":[130],"trajectories":[133],"in":[134],"principal":[136],"component":[137],"subspace.Expert's":[138],"contributions":[139],"are":[140],"based":[141],"their":[143],"projected":[144],"proximity":[145],"domains.Extensive":[148],"experiments":[149],"7":[151],"benchmarks,":[153],"5":[154],"models,":[155],"2":[157],"settings":[159],"shows":[160],"TEA":[161],"outperforms":[162],"prior":[163],"up":[167,172],"69%":[169],"being":[171],"60x":[174],"more":[175],"efficient":[176],"1":[177],".":[178]},"counts_by_year":[],"updated_date":"2026-03-12T06:13:28.667946","created_date":"2025-11-08T00:00:00"}