{"id":"https://openalex.org/W7134822109","doi":"https://doi.org/10.48550/arxiv.2603.08174","title":"MERLIN: Building Low-SNR Robust Multimodal LLMs for Electromagnetic Signals","display_name":"MERLIN: Building Low-SNR Robust Multimodal LLMs for Electromagnetic Signals","publication_year":2026,"publication_date":"2026-03-09","ids":{"openalex":"https://openalex.org/W7134822109","doi":"https://doi.org/10.48550/arxiv.2603.08174"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2603.08174","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128688498","display_name":"Junyu Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Shen, Junyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128637440","display_name":"Zhendong She","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"She, Zhendong","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111016931","display_name":"Chenghanyu Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Chenghanyu","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5088893614","display_name":"Yuchuang Sun","orcid":"https://orcid.org/0000-0001-8999-5515"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Yuchuang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128641967","display_name":"Luqing Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Luqing","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128632346","display_name":"Dingwei Tan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tan, Dingwei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128660359","display_name":"Zonghao Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Zonghao","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5040607477","display_name":"Bo Guo","orcid":"https://orcid.org/0000-0001-7723-6984"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Bo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128653231","display_name":"Zehua Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Zehua","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5022198380","display_name":"Wupeng Xie","orcid":"https://orcid.org/0000-0002-4319-9326"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Wupeng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128645124","display_name":"Yaxin Mu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mu, Yaxin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128642973","display_name":"Peng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Peng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128648243","display_name":"Peipei Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Peipei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128658796","display_name":"Fengxiang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Fengxiang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125941167","display_name":"Yangang Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Yangang","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128636053","display_name":"Maosong Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Maosong","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":16,"corresponding_author_ids":["https://openalex.org/A5128688498"],"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/T11273","display_name":"Advanced Graph Neural Networks","score":0.21639999747276306,"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"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.21639999747276306,"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"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.13740000128746033,"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"}},{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.03880000114440918,"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/robustness","display_name":"Robustness (evolution)","score":0.7559000253677368},{"id":"https://openalex.org/keywords/blueprint","display_name":"Blueprint","score":0.6732000112533569},{"id":"https://openalex.org/keywords/benchmarking","display_name":"Benchmarking","score":0.3993000090122223},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.3935000002384186},{"id":"https://openalex.org/keywords/signal-processing","display_name":"Signal processing","score":0.3287000060081482},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.30410000681877136}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.7559000253677368},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7391999959945679},{"id":"https://openalex.org/C155911762","wikidata":"https://www.wikidata.org/wiki/Q422321","display_name":"Blueprint","level":2,"score":0.6732000112533569},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4851999878883362},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44940000772476196},{"id":"https://openalex.org/C86251818","wikidata":"https://www.wikidata.org/wiki/Q816754","display_name":"Benchmarking","level":2,"score":0.3993000090122223},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3935000002384186},{"id":"https://openalex.org/C104267543","wikidata":"https://www.wikidata.org/wiki/Q208163","display_name":"Signal processing","level":3,"score":0.3287000060081482},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.30410000681877136},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.2879999876022339},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.28690001368522644},{"id":"https://openalex.org/C2777655017","wikidata":"https://www.wikidata.org/wiki/Q1501161","display_name":"Toolbox","level":2,"score":0.2842000126838684},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.27619999647140503},{"id":"https://openalex.org/C46686674","wikidata":"https://www.wikidata.org/wiki/Q466303","display_name":"Boosting (machine learning)","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.258899986743927}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2603.08174","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Article"},{"id":"doi:10.48550/arxiv.2603.08174","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.08174","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":"pmh:doi:10.48550/arxiv.2603.08174","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"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":{"The":[0,59,78],"paradigm":[1],"of":[2,61,80,90],"Multimodal":[3],"Large":[4],"Language":[5],"Models":[6],"(MLLMs)":[7],"offers":[8],"a":[9,125,130,148,190],"promising":[10],"blueprint":[11],"for":[12,73,132],"advancing":[13],"the":[14,24,46,88,135,167,183,230],"electromagnetic":[15],"(EM)":[16],"domain.":[17,137],"However,":[18],"prevailing":[19],"approaches":[20],"often":[21],"deviate":[22],"from":[23,176],"native":[25],"MLLM":[26,47],"paradigm,":[27],"instead":[28],"using":[29],"task-specific":[30],"or":[31],"pipelined":[32],"architectures":[33],"that":[34,225],"lead":[35],"to":[36,83,115,128,139,158,178,181,197,208],"fundamental":[37],"limitations":[38],"in":[39,49,101,134,215,229,236],"model":[40,211],"performance":[41,89,117,214],"and":[42,68,86,145,161,213,232],"generalization.":[43],"Fully":[44],"realizing":[45],"potential":[48],"EM":[50,66,93,136,154],"domain":[51],"requires":[52],"overcoming":[53],"three":[54],"main":[55],"challenges:":[56],"(1)":[57],"Data.":[58],"scarcity":[60],"high-quality":[62],"datasets":[63],"with":[64,202],"paired":[65],"signals":[67],"descriptive":[69],"text":[70],"annotations":[71],"used":[72],"MLLMs":[74,133],"pre-training;":[75],"(2)":[76],"Benchmark.":[77],"absence":[79],"comprehensive":[81,169],"benchmarks":[82],"systematically":[84],"evaluate":[85],"compare":[87],"models":[91],"on":[92],"signal-to-text":[94],"tasks;":[95],"(3)":[96],"Model.":[97],"A":[98],"critical":[99,108],"fragility":[100],"low":[102],"Signal-to-Noise":[103],"Ratio":[104],"(SNR)":[105],"environments,":[106],"where":[107],"signal":[109,200],"features":[110],"can":[111],"be":[112],"obscured,":[113],"leading":[114],"significant":[116],"degradation.":[118],"To":[119],"address":[120],"these":[121],"challenges,":[122],"we":[123,143,164,187],"introduce":[124],"tripartite":[126],"contribution":[127],"establish":[129],"foundation":[131],"First,":[138],"overcome":[140],"data":[141],"scarcity,":[142],"construct":[144],"release":[146],"EM-100k,":[147],"large-scale":[149],"dataset":[150],"comprising":[151],"over":[152],"100,000":[153],"signal-text":[155],"pairs.":[156],"Second,":[157],"enable":[159],"rigorous":[160],"standardized":[162],"evaluation,":[163],"propose":[165],"EM-Bench,":[166],"most":[168],"benchmark":[170],"featuring":[171],"diverse":[172],"downstream":[173],"tasks":[174],"spanning":[175],"perception":[177],"reasoning.":[179],"Finally,":[180],"tackle":[182],"core":[184],"modeling":[185],"challenge,":[186],"present":[188],"MERLIN,":[189],"novel":[191],"training":[192],"framework":[193],"designed":[194],"not":[195],"only":[196],"align":[198],"low-level":[199],"representations":[201],"high-level":[203],"semantic":[204],"text,":[205],"but":[206],"also":[207],"explicitly":[209],"enhance":[210],"robustness":[212,235],"challenging":[216],"low-SNR":[217,237],"environments.":[218],"Comprehensive":[219],"experiments":[220],"validate":[221],"our":[222],"method,":[223],"showing":[224],"MERLIN":[226],"is":[227],"state-of-the-art":[228],"EM-Bench":[231],"exhibits":[233],"remarkable":[234],"settings.":[238]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-03-11T00:00:00"}