{"id":"https://openalex.org/W7214916128","doi":"https://doi.org/10.48550/arxiv.2609.37675","title":"LEMON-ZEST: Evolution-Informed Tokenization for Efficient Protein Language Modeling","display_name":"LEMON-ZEST: Evolution-Informed Tokenization for Efficient Protein Language Modeling","publication_year":2026,"publication_date":"2026-09-29","ids":{"openalex":"https://openalex.org/W7214916128","doi":"https://doi.org/10.48550/arxiv.2609.37675"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.37675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37675","pdf_url":"https://arxiv.org/pdf/2609.37675","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"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.37675","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5153594700","display_name":"Biswajit Banerjee","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Banerjee, Biswajit","raw_affiliation_strings":["Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090644342","display_name":"Claudia \u00c1lvarez-Carre\u00f1o","orcid":"https://orcid.org/0000-0002-1827-8946","observed_orcids":["https://orcid.org/0000-0002-1827-8946"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Carreno, Claudia Alvarez","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5067399385","display_name":"Anton S. Petrov","orcid":"https://orcid.org/0000-0003-3359-7299","observed_orcids":["https://orcid.org/0000-0003-3359-7299","https://orcid.org/0000-0003-1908-135X"]},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Petrov, Anton S.","raw_affiliation_strings":["Georgia Institute of Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology","institution_ids":["https://openalex.org/I130701444"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T12254","display_name":"Machine Learning in Bioinformatics","score":0.9603999853134155,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T12254","display_name":"Machine Learning in Bioinformatics","score":0.9603999853134155,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10044","display_name":"Protein Structure and Dynamics","score":0.027799999341368675,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12946","display_name":"Fractal and DNA sequence analysis","score":0.0032999999821186066,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/protein-language-models","display_name":"protein language models","score":0.949999988079071},{"id":"https://openalex.org/keywords/multiple-sequence-alignment","display_name":"multiple sequence alignment","score":0.6869999766349792},{"id":"https://openalex.org/keywords/conserved-domains","display_name":"conserved domains","score":0.7680000066757202},{"id":"https://openalex.org/keywords/remote-homology-detection","display_name":"remote homology detection","score":0.902999997138977},{"id":"https://openalex.org/keywords/protein-representation-learning","display_name":"protein representation learning","score":0.7009999752044678}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7644000053405762},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.7382000088691711},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5065000057220459},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.476500004529953},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.36820000410079956},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.36559998989105225},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.36480000615119934},{"id":"https://openalex.org/C79029880","wikidata":"https://www.wikidata.org/wiki/Q901755","display_name":"GenBank","level":3,"score":0.3452000021934509},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.335999995470047},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.3328999876976013},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.32760000228881836},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.30070000886917114},{"id":"https://openalex.org/C170858558","wikidata":"https://www.wikidata.org/wiki/Q1394144","display_name":"Automatic summarization","level":2,"score":0.28600001335144043},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2727999985218048},{"id":"https://openalex.org/C2778355071","wikidata":"https://www.wikidata.org/wiki/Q1933849","display_name":"Microdata (statistics)","level":4,"score":0.2678000032901764}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.37675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37675","pdf_url":"https://arxiv.org/pdf/2609.37675","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":"doi:10.48550/arxiv.2609.37675","doi":"https://doi.org/10.48550/arxiv.2609.37675","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.37675","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:2609.37675","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.37675","pdf_url":"https://arxiv.org/pdf/2609.37675","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":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214916128.pdf","grobid_xml":"https://content.openalex.org/works/W7214916128.grobid-xml"},"referenced_works_count":15,"referenced_works":["https://openalex.org/W3001279689","https://openalex.org/W3006439205","https://openalex.org/W3196318247","https://openalex.org/W4317374308","https://openalex.org/W4322494707","https://openalex.org/W4327550249","https://openalex.org/W4396730528","https://openalex.org/W4403925091","https://openalex.org/W4404447386","https://openalex.org/W4405330335","https://openalex.org/W4406440058","https://openalex.org/W4415101103","https://openalex.org/W4416378091","https://openalex.org/W7133225605","https://openalex.org/W7149332052"],"related_works":["https://openalex.org/W4401254303","https://openalex.org/W7215950433","https://openalex.org/W4399283681","https://openalex.org/W3193589100","https://openalex.org/W4413950477","https://openalex.org/W7214727131","https://openalex.org/W4387303685","https://openalex.org/W4409152738","https://openalex.org/W7167603453","https://openalex.org/W4383550741"],"abstract_inverted_index":{"Protein":[0],"Language":[1],"Models":[2],"(PLMs)":[3],"have":[4],"made":[5],"remarkable":[6],"progress":[7],"following":[8],"scaling":[9],"laws":[10],"established":[11],"in":[12],"natural":[13],"language":[14],"processing":[15],"across":[16],"sequence-":[17],"and":[18,184],"structure-based":[19],"tasks,":[20],"yet":[21],"the":[22,74,190],"potential":[23],"of":[24,52,62,93,117,129],"tokenization":[25,41,75,163],"remains":[26],"underexploited.":[27],"Unlike":[28],"human":[29],"language,":[30],"proteins":[31],"preserve":[32],"structure":[33],"despite":[34],"extensive":[35],"sequence":[36,64],"variation":[37],"a":[38,104,122,137,171],"property":[39],"standard":[40,105],"strategies":[42],"fundamentally":[43],"fail":[44],"to":[45,88,99,155],"capture.":[46],"We":[47],"introduce":[48],"ZEST":[49,66,84],"(Zoned":[50],"Encoding":[51],"Sequence":[53],"Traits),":[54],"an":[55,89],"evolution-informed":[56,162],"vocabulary":[57],"derived":[58],"from":[59,120,153],"conserved":[60],"regions":[61],"multiple":[63],"alignments.":[65],"allows":[67],"embedding":[68],"domain-level":[69],"biological":[70],"priors":[71],"directly":[72],"at":[73],"stage":[76],"rather":[77],"than":[78],"learning":[79],"them":[80],"implicitly":[81],"through":[82],"scale.":[83],"natively":[85],"compresses":[86],"sequences":[87,134],"average":[90],"token":[91],"length":[92],"4":[94],"residues,":[95],"enabling":[96],"our":[97],"model":[98,126,182],"process":[100],"4,000":[101],"residues":[102],"within":[103],"1024-token":[106],"context":[107],"window.":[108],"Building":[109],"on":[110,136],"this,":[111],"we":[112],"present":[113],"LEMON":[114,148],"(Layered":[115],"Extraction":[116],"Molecular":[118],"Ordering":[119],"Nature),":[121],"compact":[123],"200M-parameter":[124],"sequence-based":[125],"for":[127,141,166,174],"detection":[128],"remote":[130],"homology":[131],"between":[132],"protein":[133,177],"trained":[135],"single":[138],"H100":[139],"GPU":[140],"one":[142],"week.":[143],"Despite":[144],"its":[145],"modest":[146],"size,":[147],"outperforms":[149],"state-of-the-art":[150],"models":[151],"ranging":[152],"600M":[154],"3B":[156],"parameters.":[157],"Our":[158],"results":[159,185],"demonstrate":[160],"that":[161],"can":[164],"substitute":[165],"massive":[167],"parameter":[168],"scaling,":[169],"opening":[170],"new":[172],"direction":[173],"efficient,":[175],"biologically-grounded":[176],"representation":[178],"learning.":[179],"All":[180],"code,":[181],"weights,":[183],"are":[186],"publicly":[187],"available":[188],"under":[189],"MIT":[191],"license.":[192]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-01T00:00:00"}