{"id":"https://openalex.org/W7125521590","doi":"https://doi.org/10.1109/icons69015.2025.00032","title":"Quantizing Small-Scale State-Space Models for Edge AI","display_name":"Quantizing Small-Scale State-Space Models for Edge AI","publication_year":2025,"publication_date":"2025-07-29","ids":{"openalex":"https://openalex.org/W7125521590","doi":"https://doi.org/10.1109/icons69015.2025.00032"},"language":"en","primary_location":{"id":"doi:10.1109/icons69015.2025.00032","doi":"https://doi.org/10.1109/icons69015.2025.00032","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icons69015.2025.00032","pdf_url":null,"source":{"id":"https://openalex.org/S7407086288","display_name":"International Conference on Neuromorphic Systems (ICONS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Neuromorphic Systems (ICONS)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","is_traditional":true,"indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2506.12480","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101504695","display_name":"Leo Zhao","orcid":"https://orcid.org/0000-0002-6688-2339","observed_orcids":["https://orcid.org/0000-0002-6688-2339"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I4405256566","display_name":"Institute of Neuroinformatics","ror":"https://ror.org/04a7xee56","country_code":"CH","type":"facility","lineage":["https://openalex.org/I202697423","https://openalex.org/I2799323385","https://openalex.org/I35440088","https://openalex.org/I4405256566"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Leo Zhao","raw_affiliation_strings":["University of Zürich and ETH,Institute of Neuroinformatics,Zürich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Zürich and ETH,Institute of Neuroinformatics,Zürich","institution_ids":["https://openalex.org/I35440088","https://openalex.org/I4405256566"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5092622432","display_name":"Tristan Torchet","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I4405256566","display_name":"Institute of Neuroinformatics","ror":"https://ror.org/04a7xee56","country_code":"CH","type":"facility","lineage":["https://openalex.org/I202697423","https://openalex.org/I2799323385","https://openalex.org/I35440088","https://openalex.org/I4405256566"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Tristan Torchet","raw_affiliation_strings":["University of Zürich and ETH,Institute of Neuroinformatics,Zürich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Zürich and ETH,Institute of Neuroinformatics,Zürich","institution_ids":["https://openalex.org/I35440088","https://openalex.org/I4405256566"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5025696765","display_name":"Melika Payvand","orcid":"https://orcid.org/0000-0001-5400-067X","observed_orcids":["https://orcid.org/0000-0001-5400-067X"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I4405256566","display_name":"Institute of Neuroinformatics","ror":"https://ror.org/04a7xee56","country_code":"CH","type":"facility","lineage":["https://openalex.org/I202697423","https://openalex.org/I2799323385","https://openalex.org/I35440088","https://openalex.org/I4405256566"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Melika Payvand","raw_affiliation_strings":["University of Zürich and ETH,Institute of Neuroinformatics,Zürich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Zürich and ETH,Institute of Neuroinformatics,Zürich","institution_ids":["https://openalex.org/I35440088","https://openalex.org/I4405256566"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070320753","display_name":"Laura Kriener","orcid":"https://orcid.org/0000-0001-5275-9199","observed_orcids":["https://orcid.org/0000-0001-5275-9199"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I4405256566","display_name":"Institute of Neuroinformatics","ror":"https://ror.org/04a7xee56","country_code":"CH","type":"facility","lineage":["https://openalex.org/I202697423","https://openalex.org/I2799323385","https://openalex.org/I35440088","https://openalex.org/I4405256566"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Laura Kriener","raw_affiliation_strings":["University of Zürich and ETH,Institute of Neuroinformatics,Zürich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Zürich and ETH,Institute of Neuroinformatics,Zürich","institution_ids":["https://openalex.org/I35440088","https://openalex.org/I4405256566"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5051201056","display_name":"Filippo Moro","orcid":"https://orcid.org/0000-0002-5279-6309","observed_orcids":["https://orcid.org/0000-0002-5279-6309","https://orcid.org/0000-0002-8922-9658"]},"institutions":[{"id":"https://openalex.org/I35440088","display_name":"ETH Zurich","ror":"https://ror.org/05a28rw58","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I35440088"]},{"id":"https://openalex.org/I4405256566","display_name":"Institute of Neuroinformatics","ror":"https://ror.org/04a7xee56","country_code":"CH","type":"facility","lineage":["https://openalex.org/I202697423","https://openalex.org/I2799323385","https://openalex.org/I35440088","https://openalex.org/I4405256566"]}],"countries":["CH"],"is_corresponding":false,"raw_author_name":"Filippo Moro","raw_affiliation_strings":["University of Zürich and ETH,Institute of Neuroinformatics,Zürich"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Zürich and ETH,Institute of Neuroinformatics,Zürich","institution_ids":["https://openalex.org/I35440088","https://openalex.org/I4405256566"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.4318,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.88263589,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":93,"max":96},"biblio":{"volume":null,"issue":null,"first_page":"158","last_page":"165"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.9829000234603882,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9829000234603882,"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/T10502","display_name":"Advanced Memory and Neural Computing","score":0.0017000000225380063,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10363","display_name":"Low-power high-performance VLSI design","score":0.0017000000225380063,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/state-space-models","display_name":"state-space model","score":0.996999979019165},{"id":"https://openalex.org/keywords/edge-intelligence","display_name":"edge AI","score":0.7540000081062317},{"id":"https://openalex.org/keywords/quantization","display_name":"quantization","score":0.5920000076293945},{"id":"https://openalex.org/keywords/post-training-quantization","display_name":"post-training quantization","score":0.847000002861023},{"id":"https://openalex.org/keywords/quantization-aware-training","display_name":"quantization-aware training","score":0.9990000128746033},{"id":"https://openalex.org/keywords/memory-efficiency","display_name":"memory efficiency","score":0.4309999942779541},{"id":"https://openalex.org/keywords/low-bit-precision","display_name":"low-bit precision","score":0.6190000176429749},{"id":"https://openalex.org/keywords/parameterization-schemes","display_name":"parameterization schemes","score":0.6389999985694885},{"id":"https://openalex.org/keywords/resource-constrained-deployment","display_name":"resource-constrained deployment","score":0.8529999852180481}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4496999979019165},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.43140000104904175},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.3379000127315521},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.33180001378059387},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3224000036716461},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2913999855518341},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2793999910354614},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2619999945163727},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.259799987077713},{"id":"https://openalex.org/C193536780","wikidata":"https://www.wikidata.org/wiki/Q1513153","display_name":"Edge detection","level":4,"score":0.2578999996185303},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.2531000077724457},{"id":"https://openalex.org/C116834253","wikidata":"https://www.wikidata.org/wiki/Q2039217","display_name":"Identification (biology)","level":2,"score":0.2529999911785126}],"mesh":[],"locations_count":4,"locations":[{"id":"doi:10.1109/icons69015.2025.00032","doi":"https://doi.org/10.1109/icons69015.2025.00032","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icons69015.2025.00032","pdf_url":null,"source":{"id":"https://openalex.org/S7407086288","display_name":"International Conference on Neuromorphic Systems (ICONS)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 International Conference on Neuromorphic Systems (ICONS)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2506.12480","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.12480","pdf_url":"https://arxiv.org/pdf/2506.12480","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:oai:www.zora.uzh.ch:20.500.14742/243091","doi":null,"is_oa":true,"landing_page_url":"https://www.zora.uzh.ch/handle/20.500.14742/243091","pdf_url":null,"source":{"id":"https://openalex.org/S4306401281","display_name":"Zurich Open Repository and Archive (University of Zurich)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I202697423","host_organization_name":"University of Zurich","host_organization_lineage":["https://openalex.org/I202697423"],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"Zhao, L., Torchet, T., Payvand, M., Kriener, L., & Moro, F. (2025). Quantizing Small-Scale State-Space Models for Edge AI. 158\u2013165. https://doi.org/10.1109/icons69015.2025.00032","raw_type":"Conference or Workshop Item"},{"id":"doi:10.48550/arxiv.2506.12480","doi":"https://doi.org/10.48550/arxiv.2506.12480","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2506.12480","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2506.12480","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2506.12480","pdf_url":"https://arxiv.org/pdf/2506.12480","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G3388106385","display_name":"UNITE: Brain-inspired device-circuits-algorithm co-design for resource-constrained hardware on the edge","funder_award_id":"211461","funder_id":"https://openalex.org/F4320320924","funder_display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung"}],"funders":[{"id":"https://openalex.org/F4320320924","display_name":"Schweizerischer Nationalfonds zur F\u00f6rderung der Wissenschaftlichen Forschung","ror":"https://ror.org/00yjd3n13"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7125521590.pdf","grobid_xml":"https://content.openalex.org/works/W7125521590.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W2777406049","https://openalex.org/W3158216587","https://openalex.org/W3213520599","https://openalex.org/W4319300630","https://openalex.org/W4377089394","https://openalex.org/W4392904222","https://openalex.org/W4402112499","https://openalex.org/W4402713424","https://openalex.org/W4404916036","https://openalex.org/W4408799708","https://openalex.org/W4412109179"],"related_works":["https://openalex.org/W4415070539","https://openalex.org/W7134919038","https://openalex.org/W7162627291","https://openalex.org/W4415966543","https://openalex.org/W4401558540","https://openalex.org/W7204914873","https://openalex.org/W7216785610","https://openalex.org/W7128642633","https://openalex.org/W7125496342","https://openalex.org/W7211982219"],"abstract_inverted_index":{"State-space":[0],"models":[1],"(SSMs)":[2],"have":[3],"recently":[4],"gained":[5],"attention":[6],"in":[7,91,133,182],"deep":[8],"learning":[9],"for":[10,22,142,178],"their":[11],"ability":[12],"to":[13,74,86,116,157],"efficiently":[14],"model":[15,158],"long-range":[16],"dependencies,":[17],"making":[18],"them":[19],"promising":[20],"candidates":[21],"edge-AI":[23],"applications.":[24],"In":[25],"this":[26],"paper,":[27],"we":[28,54,77,105,146],"analyze":[29,78],"the":[30,51,63,79,87,92,97,119,129,161],"effects":[31],"of":[32,81,131,168],"quantization":[33,58,83,150],"on":[34,40,118],"small-scale":[35],"SSMs":[36,181],"with":[37],"a":[38,148,166],"focus":[39],"reducing":[41,160],"memory":[42,163],"and":[43,60,67,89,137],"computational":[44],"costs":[45],"while":[46],"maintaining":[47],"task":[48],"performance.":[49,172],"Using":[50],"S4D":[52,93],"architecture,":[53],"first":[55],"investigate":[56],"post-training":[57],"(PTQ)":[59,115],"show":[61],"that":[62,152],"state":[64,69],"matrix":[65],"A":[66],"internal":[68],"x":[70],"are":[71],"particularly":[72],"sensitive":[73],"quantization.":[75],"Furthermore,":[76],"impact":[80],"different":[82,139,154],"techniques":[84],"applied":[85],"parameters":[88],"activations":[90],"architecture.":[94],"To":[95],"address":[96],"observed":[98],"performance":[99,112],"drop":[100],"after":[101],"Post-training":[102],"Quantization":[103],"(PTQ),":[104],"apply":[106],"Quantization-aware":[107],"Training":[108],"(QAT),":[109],"significantly":[110],"improving":[111],"from":[113],"40%":[114],"96%":[117],"sequential":[120],"MNIST":[121],"benchmark":[122],"at":[123],"8-bit":[124],"precision.":[125],"We":[126],"further":[127],"demonstrate":[128],"potential":[130],"QAT":[132,143],"enabling":[134],"sub-8-bit":[135],"precisions":[136],"evaluate":[138],"parameterization":[140],"schemes":[141],"stability.":[144],"Additionally,":[145],"propose":[147],"heterogeneous":[149],"strategy":[151],"assigns":[153],"precision":[155],"levels":[156],"components,":[159],"overall":[162],"footprint":[164],"by":[165],"factor":[167],"6x":[169],"without":[170],"sacrificing":[171],"Our":[173],"results":[174],"provide":[175],"actionable":[176],"insights":[177],"deploying":[179],"quantized":[180],"resource-constrained":[183],"environments.":[184]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2}],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-01-24T00:00:00"}