{"id":"https://openalex.org/W7162447535","doi":"https://doi.org/10.48550/arxiv.2605.25949","title":"Small Models, Strong Priors: Architectural Inductive Bias for Parameter-Efficient Neural PDE Solvers","display_name":"Small Models, Strong Priors: Architectural Inductive Bias for Parameter-Efficient Neural PDE Solvers","publication_year":2026,"publication_date":"2026-05-25","ids":{"openalex":"https://openalex.org/W7162447535","doi":"https://doi.org/10.48550/arxiv.2605.25949"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.25949","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.25949","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2605.25949","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5082927842","display_name":"Shyam Sankaran","orcid":"https://orcid.org/0000-0002-1583-2147"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sankaran, Shyam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137073303","display_name":"Hanwen Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hanwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137055471","display_name":"Paris Perdikaris","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Perdikaris, Paris","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.30469998717308044,"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":0.30469998717308044,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.1062999963760376,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.08630000054836273,"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/inductive-bias","display_name":"Inductive bias","score":0.5307999849319458},{"id":"https://openalex.org/keywords/chaotic","display_name":"Chaotic","score":0.47940000891685486},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.47540000081062317},{"id":"https://openalex.org/keywords/prior-probability","display_name":"Prior probability","score":0.44909998774528503},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.43950000405311584},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.40389999747276306},{"id":"https://openalex.org/keywords/wavelet","display_name":"Wavelet","score":0.4000999927520752},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.3984000086784363},{"id":"https://openalex.org/keywords/a-priori-and-a-posteriori","display_name":"A priori and a posteriori","score":0.3479999899864197},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.3411000072956085}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5735999941825867},{"id":"https://openalex.org/C197352929","wikidata":"https://www.wikidata.org/wiki/Q1074074","display_name":"Inductive bias","level":4,"score":0.5307999849319458},{"id":"https://openalex.org/C2777052490","wikidata":"https://www.wikidata.org/wiki/Q5072826","display_name":"Chaotic","level":2,"score":0.47940000891685486},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.47540000081062317},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.44909998774528503},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.43950000405311584},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43619999289512634},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40470001101493835},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.40389999747276306},{"id":"https://openalex.org/C47432892","wikidata":"https://www.wikidata.org/wiki/Q831390","display_name":"Wavelet","level":2,"score":0.4000999927520752},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.3984000086784363},{"id":"https://openalex.org/C75553542","wikidata":"https://www.wikidata.org/wiki/Q178161","display_name":"A priori and a posteriori","level":2,"score":0.3479999899864197},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3411000072956085},{"id":"https://openalex.org/C3770464","wikidata":"https://www.wikidata.org/wiki/Q775963","display_name":"Smoothing","level":2,"score":0.3228999972343445},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.31700000166893005},{"id":"https://openalex.org/C190839683","wikidata":"https://www.wikidata.org/wiki/Q2448197","display_name":"Train","level":2,"score":0.31360000371932983},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.30809998512268066},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.30329999327659607},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.3012000024318695},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.29809999465942383},{"id":"https://openalex.org/C2780297707","wikidata":"https://www.wikidata.org/wiki/Q4895393","display_name":"Landmark","level":2,"score":0.29789999127388},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.29490000009536743},{"id":"https://openalex.org/C2779843651","wikidata":"https://www.wikidata.org/wiki/Q7390335","display_name":"SIGNAL (programming language)","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C79379906","wikidata":"https://www.wikidata.org/wiki/Q3174497","display_name":"Dynamical systems theory","level":2,"score":0.2922999858856201},{"id":"https://openalex.org/C117251300","wikidata":"https://www.wikidata.org/wiki/Q1849855","display_name":"Parametric statistics","level":2,"score":0.2903999984264374},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.28529998660087585},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.27630001306533813},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.2750000059604645},{"id":"https://openalex.org/C35651441","wikidata":"https://www.wikidata.org/wiki/Q625303","display_name":"Independence (probability theory)","level":2,"score":0.2734000086784363},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.26030001044273376},{"id":"https://openalex.org/C192065140","wikidata":"https://www.wikidata.org/wiki/Q1147928","display_name":"Multinomial distribution","level":2,"score":0.25529998540878296}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.25949","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.25949","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2605.25949","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.25949","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Neural":[0],"PDE":[1,177],"solvers":[2],"have":[3],"followed":[4],"the":[5,42,106,115,119,153,157,190],"scaling":[6],"trajectory":[7],"of":[8,18,44,97,192],"vision":[9],"and":[10,41,48,84,111,123,188],"language,":[11],"with":[12,94,105],"recent":[13],"foundation":[14,95,141],"models":[15,92,96],"reaching":[16],"billions":[17],"parameters.":[19],"We":[20,57],"argue":[21],"that":[22,175,189],"scale":[23],"is":[24,50,179,196],"a":[25,66,79,85,139,144,169,193,197],"poor":[26],"substitute":[27],"for":[28,70],"architectural":[29,182],"inductive":[30,183],"bias":[31,184],"in":[32,61],"this":[33,59],"domain:":[34],"structured":[35],"priors":[36],"deliver":[37],"outsized":[38],"parameter":[39],"efficiency,":[40],"pattern":[43,149],"where":[45,114,152],"they":[46,55],"succeed":[47],"fail":[49],"itself":[51],"informative":[52],"about":[53,201],"what":[54],"capture.":[56],"instantiate":[58],"argument":[60],"WaveLiT,":[62],"an":[63,74],"architecture":[64],"combining":[65],"discrete":[67],"wavelet":[68],"transform":[69],"lossless":[71],"multi-resolution":[72],"tokenization,":[73],"augmented":[75],"linear":[76],"attention":[77],"block,":[78],"shared-weight":[80],"multiscale":[81],"feature":[82],"pyramid,":[83],"wavelet-domain":[86],"auxiliary":[87],"loss.":[88],"Bespoke":[89],"1-10M-parameter":[90],"WaveLiT":[91],"compete":[93],"100-1000$\\times$":[98],"their":[99],"size":[100],"across":[101,135],"eight":[102,137],"TheWell":[103],"benchmarks,":[104,138],"largest":[107],"gains":[108],"on":[109,160,168],"wave":[110],"acoustic-dominated":[112],"benchmarks":[113],"wavelet-multiscale":[116,154],"prior":[117,155],"fits":[118],"dominant":[120],"dynamical":[121],"structure":[122,191],"small":[124],"per-step":[125],"errors":[126],"do":[127],"not":[128],"compound":[129],"geometrically":[130],"under":[131],"rollout.":[132],"Trained":[133],"jointly":[134],"all":[136],"10M-parameter":[140],"variant":[142],"exhibits":[143],"structured,":[145],"physically":[146],"interpretable":[147],"transfer":[148],"--":[150],"strongest":[151],"matches":[156],"dynamics,":[158],"weakest":[159],"chaotic":[161],"advection-dominated":[162],"flows.":[163],"The":[164,172],"entire":[165],"pipeline":[166],"trains":[167],"single":[170],"GPU.":[171],"results":[173],"suggest":[174],"small-model":[176],"performance":[178],"shaped":[180],"by":[181],"rather":[185],"than":[186],"scale,":[187],"prior's":[194],"failures":[195],"useful":[198],"empirical":[199],"signal":[200],"its":[202],"content.":[203]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-27T00:00:00"}