SpectralX: Parameter-efficient Domain Generalization for Spectral Remote Sensing Foundation Models Article Swipe
Wei Li
,
Mengmeng Zhang
,
Jiawei Han
,
Ran Tao
,
Shunlin Liang
·
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2508.01731
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2508.01731
Recent advances in Remote Sensing Foundation Models (RSFMs) have led to significant breakthroughs in the field. While many RSFMs have been pretrained with massive optical imagery, more multispectral/hyperspectral data remain lack of the corresponding foundation models. To leverage the advantages of spectral imagery in earth observation, we explore whether existing RSFMs can be effectively adapted to process diverse spectral modalities without requiring extensive spectral pretraining. In response to this challenge, we proposed SpectralX, an innovative parameter-efficient fine-tuning framework that adapt existing RSFMs as backbone while introducing a two-stage training approach to handle various spectral inputs, thereby significantly improving domain generalization performance. In the first stage, we employ a masked-reconstruction task and design a specialized Hyper Tokenizer (HyperT) to extract attribute tokens from both spatial and spectral dimensions. Simultaneously, we develop an Attribute-oriented Mixture of Adapter (AoMoA) that dynamically aggregates multi-attribute expert knowledge while performing layer-wise fine-tuning. With semantic segmentation as downstream task in the second stage, we insert an Attribute-refined Adapter (Are-adapter) into the first stage framework. By iteratively querying low-level semantic features with high-level representations, the model learns to focus on task-beneficial attributes, enabling customized adjustment of RSFMs. Following this two-phase adaptation process, SpectralX is capable of interpreting spectral imagery from new regions or seasons. The codes will be available from the website: https://github.com/YuxiangZhang-BIT.
Related Topics
Concepts
No concepts available.
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2508.01731
- https://arxiv.org/pdf/2508.01731
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4417100279
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4417100279Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2508.01731Digital Object Identifier
- Title
-
SpectralX: Parameter-efficient Domain Generalization for Spectral Remote Sensing Foundation ModelsWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-08-03Full publication date if available
- Authors
-
Wei Li, Mengmeng Zhang, Jiawei Han, Ran Tao, Shunlin LiangList of authors in order
- Landing page
-
https://arxiv.org/abs/2508.01731Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2508.01731Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2508.01731Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4417100279 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2508.01731 |
| ids.doi | https://doi.org/10.48550/arxiv.2508.01731 |
| ids.openalex | https://openalex.org/W4417100279 |
| fwci | |
| type | preprint |
| title | SpectralX: Parameter-efficient Domain Generalization for Spectral Remote Sensing Foundation Models |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2508.01731 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | |
| locations[0].pdf_url | https://arxiv.org/pdf/2508.01731 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2508.01731 |
| locations[1].id | doi:10.48550/arxiv.2508.01731 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2508.01731 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5100317994 |
| authorships[0].author.orcid | https://orcid.org/0000-0001-7015-7335 |
| authorships[0].author.display_name | Wei Li |
| authorships[0].author_position | middle |
| authorships[0].raw_author_name | Li, Wei |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5100442192 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-5724-9785 |
| authorships[1].author.display_name | Mengmeng Zhang |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Zhang, Mengmeng |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5019539533 |
| authorships[2].author.orcid | https://orcid.org/0000-0002-3629-2696 |
| authorships[2].author.display_name | Jiawei Han |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Han, Jiawei |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5067803447 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-5243-7189 |
| authorships[3].author.display_name | Ran Tao |
| authorships[3].author_position | last |
| authorships[3].raw_author_name | Tao, Ran |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5054276054 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-2708-9183 |
| authorships[4].author.display_name | Shunlin Liang |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Liang, Shunlin |
| authorships[4].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2508.01731 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | SpectralX: Parameter-efficient Domain Generalization for Spectral Remote Sensing Foundation Models |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-12-08T09:52:01.160171 |
| primary_topic | |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2508.01731 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2508.01731 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2508.01731 |
| primary_location.id | pmh:oai:arXiv.org:2508.01731 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | |
| primary_location.pdf_url | https://arxiv.org/pdf/2508.01731 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2508.01731 |
| publication_date | 2025-08-03 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 86, 107, 112 |
| abstract_inverted_index.By | 167 |
| abstract_inverted_index.In | 65, 101 |
| abstract_inverted_index.To | 36 |
| abstract_inverted_index.an | 73, 130, 158 |
| abstract_inverted_index.as | 82, 149 |
| abstract_inverted_index.be | 52, 209 |
| abstract_inverted_index.in | 2, 13, 43, 152 |
| abstract_inverted_index.is | 195 |
| abstract_inverted_index.of | 31, 40, 133, 187, 197 |
| abstract_inverted_index.on | 181 |
| abstract_inverted_index.or | 204 |
| abstract_inverted_index.to | 10, 55, 67, 90, 117, 179 |
| abstract_inverted_index.we | 46, 70, 105, 128, 156 |
| abstract_inverted_index.The | 206 |
| abstract_inverted_index.and | 110, 124 |
| abstract_inverted_index.can | 51 |
| abstract_inverted_index.led | 9 |
| abstract_inverted_index.new | 202 |
| abstract_inverted_index.the | 14, 32, 38, 102, 153, 163, 176, 212 |
| abstract_inverted_index.With | 146 |
| abstract_inverted_index.been | 20 |
| abstract_inverted_index.both | 122 |
| abstract_inverted_index.data | 28 |
| abstract_inverted_index.from | 121, 201, 211 |
| abstract_inverted_index.have | 8, 19 |
| abstract_inverted_index.into | 162 |
| abstract_inverted_index.lack | 30 |
| abstract_inverted_index.many | 17 |
| abstract_inverted_index.more | 26 |
| abstract_inverted_index.task | 109, 151 |
| abstract_inverted_index.that | 78, 136 |
| abstract_inverted_index.this | 68, 190 |
| abstract_inverted_index.will | 208 |
| abstract_inverted_index.with | 22, 173 |
| abstract_inverted_index.Hyper | 114 |
| abstract_inverted_index.RSFMs | 18, 50, 81 |
| abstract_inverted_index.While | 16 |
| abstract_inverted_index.adapt | 79 |
| abstract_inverted_index.codes | 207 |
| abstract_inverted_index.earth | 44 |
| abstract_inverted_index.first | 103, 164 |
| abstract_inverted_index.focus | 180 |
| abstract_inverted_index.model | 177 |
| abstract_inverted_index.stage | 165 |
| abstract_inverted_index.while | 84, 142 |
| abstract_inverted_index.Models | 6 |
| abstract_inverted_index.RSFMs. | 188 |
| abstract_inverted_index.Recent | 0 |
| abstract_inverted_index.Remote | 3 |
| abstract_inverted_index.design | 111 |
| abstract_inverted_index.domain | 98 |
| abstract_inverted_index.employ | 106 |
| abstract_inverted_index.expert | 140 |
| abstract_inverted_index.field. | 15 |
| abstract_inverted_index.handle | 91 |
| abstract_inverted_index.insert | 157 |
| abstract_inverted_index.learns | 178 |
| abstract_inverted_index.remain | 29 |
| abstract_inverted_index.second | 154 |
| abstract_inverted_index.stage, | 104, 155 |
| abstract_inverted_index.tokens | 120 |
| abstract_inverted_index.(AoMoA) | 135 |
| abstract_inverted_index.(RSFMs) | 7 |
| abstract_inverted_index.Adapter | 134, 160 |
| abstract_inverted_index.Mixture | 132 |
| abstract_inverted_index.Sensing | 4 |
| abstract_inverted_index.adapted | 54 |
| abstract_inverted_index.capable | 196 |
| abstract_inverted_index.develop | 129 |
| abstract_inverted_index.diverse | 57 |
| abstract_inverted_index.explore | 47 |
| abstract_inverted_index.extract | 118 |
| abstract_inverted_index.imagery | 42, 200 |
| abstract_inverted_index.inputs, | 94 |
| abstract_inverted_index.massive | 23 |
| abstract_inverted_index.models. | 35 |
| abstract_inverted_index.optical | 24 |
| abstract_inverted_index.process | 56 |
| abstract_inverted_index.regions | 203 |
| abstract_inverted_index.spatial | 123 |
| abstract_inverted_index.thereby | 95 |
| abstract_inverted_index.various | 92 |
| abstract_inverted_index.whether | 48 |
| abstract_inverted_index.without | 60 |
| abstract_inverted_index.(HyperT) | 116 |
| abstract_inverted_index.advances | 1 |
| abstract_inverted_index.approach | 89 |
| abstract_inverted_index.backbone | 83 |
| abstract_inverted_index.enabling | 184 |
| abstract_inverted_index.existing | 49, 80 |
| abstract_inverted_index.features | 172 |
| abstract_inverted_index.imagery, | 25 |
| abstract_inverted_index.leverage | 37 |
| abstract_inverted_index.process, | 193 |
| abstract_inverted_index.proposed | 71 |
| abstract_inverted_index.querying | 169 |
| abstract_inverted_index.response | 66 |
| abstract_inverted_index.seasons. | 205 |
| abstract_inverted_index.semantic | 147, 171 |
| abstract_inverted_index.spectral | 41, 58, 63, 93, 125, 199 |
| abstract_inverted_index.training | 88 |
| abstract_inverted_index.website: | 213 |
| abstract_inverted_index.Following | 189 |
| abstract_inverted_index.SpectralX | 194 |
| abstract_inverted_index.Tokenizer | 115 |
| abstract_inverted_index.attribute | 119 |
| abstract_inverted_index.available | 210 |
| abstract_inverted_index.extensive | 62 |
| abstract_inverted_index.framework | 77 |
| abstract_inverted_index.improving | 97 |
| abstract_inverted_index.knowledge | 141 |
| abstract_inverted_index.low-level | 170 |
| abstract_inverted_index.requiring | 61 |
| abstract_inverted_index.two-phase | 191 |
| abstract_inverted_index.two-stage | 87 |
| abstract_inverted_index.Foundation | 5 |
| abstract_inverted_index.SpectralX, | 72 |
| abstract_inverted_index.adaptation | 192 |
| abstract_inverted_index.adjustment | 186 |
| abstract_inverted_index.advantages | 39 |
| abstract_inverted_index.aggregates | 138 |
| abstract_inverted_index.challenge, | 69 |
| abstract_inverted_index.customized | 185 |
| abstract_inverted_index.downstream | 150 |
| abstract_inverted_index.foundation | 34 |
| abstract_inverted_index.framework. | 166 |
| abstract_inverted_index.high-level | 174 |
| abstract_inverted_index.innovative | 74 |
| abstract_inverted_index.layer-wise | 144 |
| abstract_inverted_index.modalities | 59 |
| abstract_inverted_index.performing | 143 |
| abstract_inverted_index.pretrained | 21 |
| abstract_inverted_index.attributes, | 183 |
| abstract_inverted_index.dimensions. | 126 |
| abstract_inverted_index.dynamically | 137 |
| abstract_inverted_index.effectively | 53 |
| abstract_inverted_index.fine-tuning | 76 |
| abstract_inverted_index.introducing | 85 |
| abstract_inverted_index.iteratively | 168 |
| abstract_inverted_index.significant | 11 |
| abstract_inverted_index.specialized | 113 |
| abstract_inverted_index.fine-tuning. | 145 |
| abstract_inverted_index.interpreting | 198 |
| abstract_inverted_index.observation, | 45 |
| abstract_inverted_index.performance. | 100 |
| abstract_inverted_index.pretraining. | 64 |
| abstract_inverted_index.segmentation | 148 |
| abstract_inverted_index.(Are-adapter) | 161 |
| abstract_inverted_index.breakthroughs | 12 |
| abstract_inverted_index.corresponding | 33 |
| abstract_inverted_index.significantly | 96 |
| abstract_inverted_index.generalization | 99 |
| abstract_inverted_index.Simultaneously, | 127 |
| abstract_inverted_index.multi-attribute | 139 |
| abstract_inverted_index.task-beneficial | 182 |
| abstract_inverted_index.representations, | 175 |
| abstract_inverted_index.Attribute-refined | 159 |
| abstract_inverted_index.Attribute-oriented | 131 |
| abstract_inverted_index.parameter-efficient | 75 |
| abstract_inverted_index.masked-reconstruction | 108 |
| abstract_inverted_index.multispectral/hyperspectral | 27 |
| abstract_inverted_index.https://github.com/YuxiangZhang-BIT. | 214 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 5 |
| citation_normalized_percentile |