{"id":"https://openalex.org/W7204452191","doi":"https://doi.org/10.48550/arxiv.2608.25128","title":"When Does Context Routing Help? A Systematic Study of Multi-Modal Fusion in Time Series Forecasting","display_name":"When Does Context Routing Help? A Systematic Study of Multi-Modal Fusion in Time Series Forecasting","publication_year":2026,"publication_date":"2026-08-25","ids":{"openalex":"https://openalex.org/W7204452191","doi":"https://doi.org/10.48550/arxiv.2608.25128"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.25128","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25128","pdf_url":"https://arxiv.org/pdf/2608.25128","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.25128","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5148424743","display_name":"Ruizhe Zhou","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Ruizhe","raw_affiliation_strings":["Seattle , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seattle , USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5061980801","display_name":"Gaoyuan Du","orcid":"https://orcid.org/0000-0002-2374-186X","observed_orcids":["https://orcid.org/0000-0002-2374-186X"]},"institutions":[{"id":"https://openalex.org/I75027704","display_name":"University of Tennessee at Knoxville","ror":"https://ror.org/020f3ap87","country_code":"US","type":"education","lineage":["https://openalex.org/I75027704"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Du, Gaoyuan","raw_affiliation_strings":["University of Tennessee , Knoxville Knoxville , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Tennessee , Knoxville Knoxville , USA","institution_ids":["https://openalex.org/I75027704"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137032959","display_name":"Xiaoyang Liu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xiaoyang","raw_affiliation_strings":["Seattle , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seattle , USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5148440203","display_name":"Haoqi Yao","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Haoqi","raw_affiliation_strings":["WorkMagic New York , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"WorkMagic New York , USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078048346","display_name":"Deepayan Chakrabarti","orcid":"https://orcid.org/0000-0002-3863-4928","observed_orcids":["https://orcid.org/0000-0002-3863-4928"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chakrabarti, Deepayan","raw_affiliation_strings":["University of Texas at Austin Austin , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Texas at Austin Austin , USA","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5012529481","display_name":"Juchun Lin","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Jiating","raw_affiliation_strings":["Seattle , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Seattle , USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5148424319","display_name":"Yixuan Shen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Yixuan","raw_affiliation_strings":["Chicago , USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chicago , USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T12205","display_name":"Time Series Analysis and Forecasting","score":0.8988999724388123,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T12205","display_name":"Time Series Analysis and Forecasting","score":0.8988999724388123,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T11918","display_name":"Forecasting Techniques and Applications","score":0.09080000221729279,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T11326","display_name":"Stock Market Forecasting Methods","score":0.004000000189989805,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/multimodal-fusion","display_name":"multimodal fusion","score":0.7289999723434448},{"id":"https://openalex.org/keywords/mixture-of-experts","display_name":"mixture of experts","score":0.7480000257492065},{"id":"https://openalex.org/keywords/conditional-mutual-information","display_name":"conditional mutual information","score":0.7210000157356262},{"id":"https://openalex.org/keywords/autocorrelation","display_name":"autocorrelation","score":0.7509999871253967},{"id":"https://openalex.org/keywords/shortcut-learning","display_name":"shortcut learning","score":0.8410000205039978},{"id":"https://openalex.org/keywords/causal-intervention","display_name":"causal intervention","score":0.4020000100135803},{"id":"https://openalex.org/keywords/quality-of-context","display_name":"quality of context","score":0.5189999938011169}],"concepts":[{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.7796000242233276},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7215999960899353},{"id":"https://openalex.org/C5297727","wikidata":"https://www.wikidata.org/wiki/Q786970","display_name":"Autocorrelation","level":2,"score":0.6061999797821045},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.5164999961853027},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.5009999871253967},{"id":"https://openalex.org/C64869954","wikidata":"https://www.wikidata.org/wiki/Q1859747","display_name":"False positive paradox","level":2,"score":0.45660001039505005},{"id":"https://openalex.org/C143724316","wikidata":"https://www.wikidata.org/wiki/Q312468","display_name":"Series (stratigraphy)","level":2,"score":0.45339998602867126},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44929999113082886},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.43849998712539673},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.438400000333786},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4366999864578247},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4198000133037567},{"id":"https://openalex.org/C64357122","wikidata":"https://www.wikidata.org/wiki/Q1149766","display_name":"Causality (physics)","level":2,"score":0.4000000059604645},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.36570000648498535},{"id":"https://openalex.org/C31395832","wikidata":"https://www.wikidata.org/wiki/Q1318674","display_name":"Testbed","level":2,"score":0.3621000051498413},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.358599990606308},{"id":"https://openalex.org/C151406439","wikidata":"https://www.wikidata.org/wiki/Q186588","display_name":"Time series","level":2,"score":0.34860000014305115},{"id":"https://openalex.org/C33954974","wikidata":"https://www.wikidata.org/wiki/Q486494","display_name":"Sensor fusion","level":2,"score":0.3264999985694885},{"id":"https://openalex.org/C152565575","wikidata":"https://www.wikidata.org/wiki/Q1124538","display_name":"Conditional random field","level":2,"score":0.3028999865055084},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.29589998722076416},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2766999900341034},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.27489998936653137},{"id":"https://openalex.org/C2779662365","wikidata":"https://www.wikidata.org/wiki/Q5416694","display_name":"Event (particle physics)","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.2551000118255615}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.25128","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25128","pdf_url":"https://arxiv.org/pdf/2608.25128","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.25128","doi":"https://doi.org/10.48550/arxiv.2608.25128","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.25128","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.25128","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.25128","pdf_url":"https://arxiv.org/pdf/2608.25128","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,"listed_in":[],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204452191.pdf","grobid_xml":"https://content.openalex.org/works/W7204452191.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4403579092","https://openalex.org/W7222997021","https://openalex.org/W7213340110","https://openalex.org/W7204965436","https://openalex.org/W7136130104","https://openalex.org/W7203789329","https://openalex.org/W7126256145","https://openalex.org/W7167852552","https://openalex.org/W7213922448","https://openalex.org/W4388186283"],"abstract_inverted_index":{"Multi-modal":[0],"time":[1],"series":[2],"forecasting":[3],"methods":[4],"integrate":[5],"auxiliary":[6,49],"context":[7,36,50,81,182],"into":[8],"temporal":[9],"predictions":[10],"through":[11,164],"increasingly":[12],"sophisticated":[13],"fusion":[14,118],"mechanisms.":[15],"A":[16],"growing":[17],"body":[18],"of":[19,34,152,197,230],"work":[20],"reports":[21],"substantial":[22],"gains,":[23],"yet":[24],"it":[25],"is":[26,68,236],"often":[27],"unclear":[28],"whether":[29],"they":[30],"reflect":[31],"genuine":[32],"use":[33],"the":[35,66,80,85,145,149,153,185,194,213,228,233,240,258],"or":[37],"incidental":[38],"architectural":[39],"effects.":[40],"We":[41,56,161,192,205,224],"ask":[42],"a":[43,52,72,122,138,168,207,246],"narrower,":[44],"checkable":[45],"question:":[46],"when":[47,94,130,142],"can":[48,98],"help":[51],"forecaster":[53],"at":[54],"all?":[55],"identify":[57],"two":[58,165],"dataset-level":[59],"conditions":[60,132],"that":[61,129],"must":[62],"both":[63,131],"hold:":[64],"(1)":[65],"target":[67,86],"not":[69],"dominated":[70],"by":[71,175,257],"last-value":[73],"shortcut":[74,169],"(low":[75],"autocorrelation":[76,195],"rho_h),":[77],"and":[78,115,156,251],"(2)":[79],"carries":[82,157],"information":[83,92],"about":[84,227],"beyond":[87],"history":[88],"(non-zero":[89],"conditional":[90],"mutual":[91],"delta;":[93],"delta=0":[95],"no":[96,158,218],"predictor":[97],"benefit---a":[99],"distribution-free":[100],"result).":[101],"Through":[102],"controlled":[103],"experiments":[104],"on":[105,200,212],"MoME":[106,171],"(a":[107],"14.3B-parameter":[108],"mixture-of-experts":[109],"model,":[110],"6":[111],"datasets,":[112],"10":[113],"seeds)":[114],"four":[116],"additional":[117],"mechanisms":[119],"implemented":[120],"within":[121],"single-backbone":[123],"testbed":[124],"(5":[125],"datasets),":[126],"we":[127,215],"find":[128],"hold,":[133],"text-conditioned":[134],"expert":[135],"modulation":[136,154],"contributes":[137],"sizeable":[139],"MSE":[140],"reduction;":[141],"either":[143],"fails,":[144],"contribution":[146,174],"collapses":[147],"to":[148,170,190],"capacity":[150],"floor":[151],"pathway":[155],"context-attributable":[159],"signal.":[160],"establish":[162],"causality":[163],"interventions:":[166],"adding":[167],"suppresses":[172],"routing":[173],"77-93%":[176],"across":[177],"3":[178],"datasets;":[179],"progressively":[180],"corrupting":[181],"quality":[183],"drives":[184],"context-specific":[186],"benefit":[187],"from":[188,245],"+44%":[189],"negative.":[191],"validate":[193],"component":[196],"our":[198,231],"diagnostic":[199,210],"27":[201],"Monash":[202],"Archive":[203],"datasets.":[204],"provide":[206],"calibrated":[208],"pre-training":[209],"that,":[211],"datasets":[214],"test,":[216],"yields":[217],"false":[219],"positives":[220],"in":[221,255],"well-powered":[222],"settings.":[223],"are":[225,252],"explicit":[226],"asymmetry":[229],"evidence:":[232],"negative":[234],"arm":[235],"broadly":[237],"reliable,":[238],"while":[239],"large":[241],"positive":[242],"magnitudes":[243],"come":[244],"single":[247],"model":[248],"family":[249],"(MoME)":[250],"corroborated":[253],"only":[254],"direction":[256],"testbed.":[259]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-28T00:00:00"}