{"id":"https://openalex.org/W7133317338","doi":"https://doi.org/10.48550/arxiv.2603.01140","title":"Teacher-Guided Causal Interventions for Image Denoising: Orthogonal Content-Noise Disentanglement in Vision Transformers","display_name":"Teacher-Guided Causal Interventions for Image Denoising: Orthogonal Content-Noise Disentanglement in Vision Transformers","publication_year":2026,"publication_date":"2026-03-01","ids":{"openalex":"https://openalex.org/W7133317338","doi":"https://doi.org/10.48550/arxiv.2603.01140"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.01140","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01140","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.01140","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127975782","display_name":"Kuai Jiang","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Jiang, Kuai","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128000527","display_name":"Zhaoyan Ding","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ding, Zhaoyan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127993695","display_name":"Guijuan Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Guijuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025153376","display_name":"Dianjie Lu","orcid":"https://orcid.org/0000-0001-5435-5307"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Dianjie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128018362","display_name":"Zhuoran Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Zhuoran","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5127975782"],"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":0.6970000267028809,"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":0.6970000267028809,"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/T10688","display_name":"Image and Signal Denoising Methods","score":0.1340000033378601,"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/T11019","display_name":"Image Enhancement Techniques","score":0.05429999902844429,"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/spurious-relationship","display_name":"Spurious relationship","score":0.6880999803543091},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6341000199317932},{"id":"https://openalex.org/keywords/subspace-topology","display_name":"Subspace topology","score":0.5037999749183655},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.4871000051498413},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.42489999532699585},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.3944000005722046},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.37209999561309814},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.35440000891685486},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.3531000018119812},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.35269999504089355}],"concepts":[{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.6880999803543091},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6341000199317932},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5982999801635742},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5266000032424927},{"id":"https://openalex.org/C32834561","wikidata":"https://www.wikidata.org/wiki/Q660730","display_name":"Subspace topology","level":2,"score":0.5037999749183655},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.4871000051498413},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.42489999532699585},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3944000005722046},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38280001282691956},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.37209999561309814},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.35440000891685486},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.3531000018119812},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.35269999504089355},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.35030001401901245},{"id":"https://openalex.org/C157202957","wikidata":"https://www.wikidata.org/wiki/Q1659609","display_name":"Image warping","level":2,"score":0.33239999413490295},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.30979999899864197},{"id":"https://openalex.org/C113364801","wikidata":"https://www.wikidata.org/wiki/Q26674","display_name":"High fidelity","level":2,"score":0.30489999055862427},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3012999892234802},{"id":"https://openalex.org/C29265498","wikidata":"https://www.wikidata.org/wiki/Q7047719","display_name":"Noise measurement","level":3,"score":0.2953000068664551},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C2776036281","wikidata":"https://www.wikidata.org/wiki/Q48769818","display_name":"Constraint (computer-aided design)","level":2,"score":0.2870999872684479},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.2849999964237213},{"id":"https://openalex.org/C18653775","wikidata":"https://www.wikidata.org/wiki/Q1333358","display_name":"Joint probability distribution","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C158600405","wikidata":"https://www.wikidata.org/wiki/Q5054566","display_name":"Causal inference","level":2,"score":0.27630001306533813},{"id":"https://openalex.org/C131720326","wikidata":"https://www.wikidata.org/wiki/Q5196075","display_name":"Curvelet","level":4,"score":0.2757999897003174},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.26510000228881836},{"id":"https://openalex.org/C35772409","wikidata":"https://www.wikidata.org/wiki/Q1323086","display_name":"Image noise","level":3,"score":0.2646999955177307},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.2621999979019165},{"id":"https://openalex.org/C2780909371","wikidata":"https://www.wikidata.org/wiki/Q4801092","display_name":"Artificial noise","level":4,"score":0.26010000705718994},{"id":"https://openalex.org/C5274069","wikidata":"https://www.wikidata.org/wiki/Q2285707","display_name":"Categorical variable","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.25290000438690186},{"id":"https://openalex.org/C11671645","wikidata":"https://www.wikidata.org/wiki/Q5054567","display_name":"Causal model","level":2,"score":0.2522999942302704},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.25040000677108765}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.01140","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01140","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.01140","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.01140","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":"article"},"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":{"Conventional":[0],"image":[1,153],"denoising":[2,41],"models":[3],"often":[4],"inadvertently":[5],"learn":[6],"spurious":[7],"correlations":[8],"between":[9,132],"environmental":[10,115],"factors":[11],"and":[12,134,184],"noise":[13,36,135],"patterns.":[14],"Moreover,":[15],"due":[16],"to":[17,22,112,127,156],"high-frequency":[18],"ambiguity,":[19,144],"they":[20],"struggle":[21],"reliably":[23],"distinguish":[24],"subtle":[25],"textures":[26],"from":[27],"stochastic":[28],"noise,":[29,55],"resulting":[30],"in":[31,181],"over-removed":[32],"details":[33],"or":[34],"residual":[35],"artifacts.":[37],"We":[38],"therefore":[39],"revisit":[40],"via":[42,80],"causal":[43,159],"intervention,":[44],"arguing":[45],"that":[46,173],"purely":[47],"correlational":[48],"fitting":[49],"entangles":[50],"intrinsic":[51],"content":[52,133,163],"with":[53],"extrinsic":[54],"which":[56,74],"directly":[57],"degrades":[58],"robustness":[59],"under":[60],"distribution":[61],"shifts.":[62],"Motivated":[63],"by":[64],"this,":[65],"we":[66,145],"propose":[67],"the":[68,77,167],"Teacher-Guided":[69],"Causal":[70],"Disentanglement":[71],"Network":[72],"(TCD-Net),":[73],"explicitly":[75],"decomposes":[76],"generative":[78],"mechanism":[79],"structured":[81],"interventions":[82],"on":[83,193],"feature":[84],"spaces":[85],"within":[86],"a":[87,108,129,158,187,194],"Vision":[88],"Transformer":[89],"framework.":[90],"Specifically,":[91],"our":[92],"method":[93],"integrates":[94],"three":[95],"key":[96],"components:":[97],"(1)":[98],"An":[99],"Environmental":[100],"Bias":[101],"Adjustment":[102],"(EBA)":[103],"module":[104],"projects":[105],"features":[106],"into":[107],"stable,":[109],"de-centered":[110],"subspace":[111],"suppress":[113],"global":[114],"bias":[116],"(de-confounding).":[117],"(2)":[118],"A":[119],"dual-branch":[120],"disentanglement":[121],"head":[122],"employs":[123],"an":[124],"orthogonality":[125],"constraint":[126],"force":[128],"strict":[130],"separation":[131],"representations,":[136],"preventing":[137],"information":[138],"leakage.":[139],"(3)":[140],"To":[141],"resolve":[142],"structural":[143],"leverage":[146],"Nano":[147],"Banana":[148],"Pro,":[149],"Google's":[150],"reasoning-guided":[151],"AI":[152],"generation":[154],"model,":[155],"guide":[157],"prior,":[160],"effectively":[161],"pulling":[162],"representations":[164],"back":[165],"onto":[166],"natural-image":[168],"manifold.":[169],"Extensive":[170],"experiments":[171],"demonstrate":[172],"TCD-Net":[174],"outperforms":[175],"mainstream":[176],"methods":[177],"across":[178],"multiple":[179],"benchmarks":[180],"both":[182],"fidelity":[183],"efficiency,":[185],"achieving":[186],"real-time":[188],"speed":[189],"of":[190],"104.2":[191],"FPS":[192],"single":[195],"RTX":[196],"5090":[197],"GPU.":[198]},"counts_by_year":[],"updated_date":"2026-03-04T07:09:34.246503","created_date":"2026-03-04T00:00:00"}