{"id":"https://openalex.org/W7164899210","doi":"https://doi.org/10.48550/arxiv.2606.16602","title":"PhysGuard: Fisher-Guided Gradient Projection for Sim-to-Real Neural PDE Surrogates","display_name":"PhysGuard: Fisher-Guided Gradient Projection for Sim-to-Real Neural PDE Surrogates","publication_year":2026,"publication_date":"2026-06-15","ids":{"openalex":"https://openalex.org/W7164899210","doi":"https://doi.org/10.48550/arxiv.2606.16602"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.16602","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16602","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":"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"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.16602","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017379162","display_name":"Changjian Zhou","orcid":"https://orcid.org/0000-0002-2094-6405"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Changjian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036603233","display_name":"Ji Fang","orcid":"https://orcid.org/0000-0003-3418-535X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Junfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020963012","display_name":"Negin Yousefpour","orcid":"https://orcid.org/0000-0002-6634-2658"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yousefpour, Negin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138707196","display_name":"Peng Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Peng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138728702","display_name":"Bin Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Bin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138738034","display_name":"Guillermo A Narsilio","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Narsilio, Guillermo A","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":0.29679998755455017,"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":0.29679998755455017,"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":0.28220000863075256,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.07599999755620956,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.555400013923645},{"id":"https://openalex.org/keywords/operator","display_name":"Operator (biology)","score":0.5260999798774719},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5249000191688538},{"id":"https://openalex.org/keywords/projection","display_name":"Projection (relational algebra)","score":0.5095000267028809},{"id":"https://openalex.org/keywords/subspace-topology","display_name":"Subspace topology","score":0.4918000102043152},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.47600001096725464},{"id":"https://openalex.org/keywords/domain","display_name":"Domain (mathematical analysis)","score":0.400299996137619},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.364300012588501},{"id":"https://openalex.org/keywords/domain-adaptation","display_name":"Domain adaptation","score":0.36419999599456787}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6787999868392944},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.555400013923645},{"id":"https://openalex.org/C17020691","wikidata":"https://www.wikidata.org/wiki/Q139677","display_name":"Operator (biology)","level":5,"score":0.5260999798774719},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5249000191688538},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.510200023651123},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.5095000267028809},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4927000105381012},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.4918000102043152},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.47600001096725464},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.400299996137619},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.364300012588501},{"id":"https://openalex.org/C2776434776","wikidata":"https://www.wikidata.org/wiki/Q19246213","display_name":"Domain adaptation","level":3,"score":0.36419999599456787},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3490000069141388},{"id":"https://openalex.org/C106487976","wikidata":"https://www.wikidata.org/wiki/Q685816","display_name":"Matrix (chemical analysis)","level":2,"score":0.33820000290870667},{"id":"https://openalex.org/C3832189","wikidata":"https://www.wikidata.org/wiki/Q8588916","display_name":"Models of neural computation","level":3,"score":0.3240000009536743},{"id":"https://openalex.org/C22789450","wikidata":"https://www.wikidata.org/wiki/Q420904","display_name":"Singular value decomposition","level":2,"score":0.32010000944137573},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3199999928474426},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.31610000133514404},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3124000132083893},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.31200000643730164},{"id":"https://openalex.org/C29406490","wikidata":"https://www.wikidata.org/wiki/Q1420659","display_name":"Fisher information","level":2,"score":0.311599999666214},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.31130000948905945},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.29649999737739563},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.2937999963760376},{"id":"https://openalex.org/C42355184","wikidata":"https://www.wikidata.org/wiki/Q1361088","display_name":"Matrix decomposition","level":3,"score":0.2705000042915344},{"id":"https://openalex.org/C2777402240","wikidata":"https://www.wikidata.org/wiki/Q6783436","display_name":"Masking (illustration)","level":2,"score":0.26980000734329224},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2639999985694885}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.16602","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16602","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.16602","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.16602","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":"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"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Neural":[0,73],"operator":[1,175],"models":[2,137],"trained":[3],"on":[4,109,170,186],"simulation":[5,110],"data":[6,25,111],"often":[7],"lose":[8],"accuracy":[9],"when":[10],"applied":[11],"to":[12,16,76,112,121,191,209,212],"experimental":[13],"measurements":[14],"due":[15],"the":[17,35,103,148,158],"sim-to-real":[18,95],"gap.":[19],"Standard":[20],"fine-tuning":[21,119,214],"with":[22,127,138,165],"limited":[23],"real":[24],"can":[26],"reduce":[27],"this":[28,134],"gap,":[29],"but":[30],"it":[31,54,203],"may":[32],"also":[33],"damage":[34],"core":[36],"physics-relevant":[37],"representations":[38],"learned":[39],"during":[40],"pretraining.":[41],"Although":[42],"knowledge-preserving":[43],"adaptation":[44,96],"has":[45],"been":[46],"widely":[47],"investigated":[48],"in":[49],"vision":[50],"or":[51,84],"language":[52],"tasks,":[53],"remains":[55],"unclear":[56],"whether":[57],"these":[58],"methods":[59],"are":[60,70,162,195],"suitable":[61],"for":[62,93,136],"neural":[63,98,174],"operators":[64,74],"whose":[65],"architectures":[66,176],"and":[67,177],"protected":[68,149],"knowledge":[69],"fundamentally":[71],"different.":[72],"need":[75],"preserve":[77],"core-scale":[78],"physical":[79,179],"structures":[80],"rather":[81],"than":[82],"semantic":[83],"visual":[85],"features.":[86],"We":[87],"propose":[88],"PhysGuard,":[89],"a":[90],"physics-preserving":[91],"framework":[92],"accurate":[94],"of":[97,140],"operators.":[99],"Specifically,":[100],"PhysGuard":[101,183],"uses":[102],"empirical":[104],"Fisher":[105,160],"Information":[106],"Matrix":[107],"computed":[108],"identify":[113],"physics-critical":[114],"parameter":[115],"directions,":[116],"then":[117],"restricts":[118],"updates":[120],"directions":[122,161],"that":[123,157,182],"do":[124],"not":[125],"interfere":[126],"them.":[128],"A":[129,152],"layer-wise":[130],"Gram-matrix":[131],"formulation":[132],"makes":[133],"efficient":[135],"millions":[139],"parameters,":[141],"while":[142,215],"an":[143],"adaptive":[144],"threshold":[145],"automatically":[146],"determines":[147],"subspace":[150],"size.":[151],"spectral":[153],"probe":[154],"experiment":[155],"shows":[156],"dominant":[159],"strongly":[163,185],"associated":[164],"low-frequency":[166,205],"output":[167],"structures.":[168],"Experiments":[169],"benchmark":[171],"across":[172],"four":[173],"different":[178],"systems":[180],"show":[181],"performs":[184],"most":[187,196],"evaluation":[188],"metrics":[189],"compared":[190,211],"baselines.":[192],"The":[193],"benefits":[194],"evident":[197],"under":[198],"severe":[199],"domain":[200],"shift,":[201],"where":[202],"reduces":[204],"error":[206],"by":[207],"up":[208],"32\\%":[210],"standard":[213],"maintaining":[216],"adaptability.":[217],"Our":[218],"code":[219],"is":[220],"available":[221],"at":[222],"https://github.com/ZhouChaunge/PhysGuard.":[223]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-17T00:00:00"}