{"id":"https://openalex.org/W6929710474","doi":"https://doi.org/10.5075/epfl-thesis-10383","title":"Discovering meaningful units from text sequences","display_name":"Discovering meaningful units from text sequences","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W6929710474","doi":"https://doi.org/10.5075/epfl-thesis-10383"},"language":"en","primary_location":{"id":"pmh:oai:infoscience.epfl.ch:20.500.14299/241742","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/241742","pdf_url":"https://infoscience.epfl.ch/bitstreams/d10b3856-0273-4271-9c52-69700d09c550/download","source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"THESIS","raw_type":"doctoral thesis"},"type":"dissertation","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://infoscience.epfl.ch/bitstreams/d10b3856-0273-4271-9c52-69700d09c550/download","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Behjati, Melika","orcid":null},"institutions":[{"id":"https://openalex.org/I5124864","display_name":"\u00c9cole Polytechnique F\u00e9d\u00e9rale de Lausanne","ror":"https://ror.org/02s376052","country_code":"CH","type":"education","lineage":["https://openalex.org/I2799323385","https://openalex.org/I5124864"]}],"countries":["CH"],"is_corresponding":true,"raw_author_name":"Behjati, Melika","raw_affiliation_strings":["Ecole polytechnique f\u00e9d\u00e9rale de Lausanne"],"affiliations":[{"raw_affiliation_string":"Ecole polytechnique f\u00e9d\u00e9rale de Lausanne","institution_ids":["https://openalex.org/I5124864"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I5124864"],"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":true,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.2903999984264374,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.2903999984264374,"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":0.15790000557899475,"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":0.10130000114440918,"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/byte","display_name":"Byte","score":0.7842000126838684},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.7229999899864197},{"id":"https://openalex.org/keywords/word","display_name":"Word (group theory)","score":0.6398000121116638},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.5746999979019165},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.531499981880188},{"id":"https://openalex.org/keywords/preprocessor","display_name":"Preprocessor","score":0.5184000134468079},{"id":"https://openalex.org/keywords/field","display_name":"Field (mathematics)","score":0.4519999921321869},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.43479999899864197}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7991999983787537},{"id":"https://openalex.org/C43364308","wikidata":"https://www.wikidata.org/wiki/Q8799","display_name":"Byte","level":2,"score":0.7842000126838684},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.7229999899864197},{"id":"https://openalex.org/C90805587","wikidata":"https://www.wikidata.org/wiki/Q10944557","display_name":"Word (group theory)","level":2,"score":0.6398000121116638},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6389999985694885},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6353999972343445},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.5746999979019165},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.531499981880188},{"id":"https://openalex.org/C34736171","wikidata":"https://www.wikidata.org/wiki/Q918333","display_name":"Preprocessor","level":2,"score":0.5184000134468079},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.4519999921321869},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.43479999899864197},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.43459999561309814},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3693000078201294},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3131999969482422},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3091000020503998},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.28859999775886536},{"id":"https://openalex.org/C35639132","wikidata":"https://www.wikidata.org/wiki/Q7452468","display_name":"Sequence labeling","level":3,"score":0.2667999863624573},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.26570001244544983},{"id":"https://openalex.org/C2780861071","wikidata":"https://www.wikidata.org/wiki/Q1062934","display_name":"Character (mathematics)","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2513999938964844}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:infoscience.epfl.ch:20.500.14299/241742","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/241742","pdf_url":"https://infoscience.epfl.ch/bitstreams/d10b3856-0273-4271-9c52-69700d09c550/download","source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"THESIS","raw_type":"doctoral thesis"},{"id":"doi:10.5075/epfl-thesis-10383","is_oa":true,"landing_page_url":"https://doi.org/10.5075/epfl-thesis-10383","pdf_url":null,"source":{"id":"https://openalex.org/S4306400488","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"thesis"}],"best_oa_location":{"id":"pmh:oai:infoscience.epfl.ch:20.500.14299/241742","is_oa":true,"landing_page_url":"https://infoscience.epfl.ch/handle/20.500.14299/241742","pdf_url":"https://infoscience.epfl.ch/bitstreams/d10b3856-0273-4271-9c52-69700d09c550/download","source":{"id":"https://openalex.org/S4306400487","display_name":"Infoscience (Ecole Polytechnique F\u00e9d\u00e9rale de Lausanne)","issn_l":null,"issn":null,"is_oa":true,"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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"THESIS","raw_type":"doctoral thesis"},"sustainable_development_goals":[{"score":0.828680694103241,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W6929710474.pdf","grobid_xml":"https://content.openalex.org/works/W6929710474.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"In":[0,40,145,156,168,271,314,345],"recent":[1],"years,":[2],"the":[3,14,34,52,64,74,76,89,93,138,175,204,236,258,266,272,315,338,346,352,373,376],"field":[4],"of":[5,16,20,23,28,66,166,178,201,260,268,286,329,341],"Natural":[6],"Language":[7],"Processing":[8],"has":[9,133,141],"seen":[10],"significant":[11],"revolution":[12],"by":[13,309,331,369],"introduction":[15],"Transformers,":[17],"a":[18,57,190,211,277,284,290,320,327,342,361,381],"stack":[19],"multiple":[21],"layers":[22,340],"attention":[24],"and":[25,33,129,173,264,355],"non-linearity,":[26],"capable":[27],"performing":[29],"almost":[30],"any":[31,61],"task":[32,59,177,237],"backbone":[35],"for":[36,207,249,300],"large":[37],"foundation":[38],"models.":[39,83,270],"contrast":[41],"to":[42,50,55,81,114,150,158,209,225,244,279,305,351,372],"traditional":[43],"NLP":[44],"pipelines,":[45],"this":[46,169,307],"architecture":[47],"is":[48,73,79,113,189,203,238],"able":[49,224],"learn":[51],"features":[53],"required":[54],"perform":[56,159,210],"specific":[58,212],"without":[60,228],"assumptions":[62],"about":[63],"structure":[65],"language.":[67],"The":[68,109],"only":[69],"remaining":[70],"hard-coded":[71],"aspect":[72],"way":[75],"text":[77,182],"input":[78,94,139],"fed":[80],"these":[82],"This":[84,187],"preprocessing":[85],"step,":[86,91],"known":[87],"as":[88,100,102,105,107,118],"tokenization":[90],"divides":[92],"into":[95],"chunks":[96],"which":[97,125,196,199,222,262,322,363],"could":[98],"be":[99],"fine-grained":[101],"bytes":[103],"or":[104,122],"coarse-grained":[106],"words.":[108,130],"most":[110,205],"popular":[111],"approach":[112,383],"use":[115],"subwords,":[116,358],"such":[117],"Byte":[119],"Pair":[120],"Encodings":[121],"word":[123],"pieces":[124],"lie":[126],"between":[127],"characters":[128,330],"However,":[131],"it":[132,147],"been":[134],"shown":[135],"that":[136],"hard-coding":[137],"representations,":[140],"its":[142,246],"own":[143],"drawbacks.":[144],"particular,":[146],"would":[148],"lead":[149],"sub-optimal":[151],"performance":[152,267],"in":[153,183,295,337,375],"downstream":[154],"tasks.":[155],"addition,":[157],"different":[160,164,231],"tasks":[161,261],"we":[162,171,219,240,252,254,275,318,349,359,379],"need":[163,241],"levels":[165],"representations.":[167],"thesis,":[170],"define":[172],"address":[174],"novel":[176,242],"inducing":[179,385],"units":[180,227,282,325],"from":[181,283,292,326,357,366],"an":[184],"unsupervised":[185],"manner.":[186],"work":[188],"step":[191],"towards":[192,384],"completely":[193],"end-to-end":[194],"models":[195,221],"can":[197],"decide":[198],"level":[200],"representation":[202],"suitable":[206],"them":[208,371],"task.":[213],"Our":[214],"contributions":[215],"are":[216,223],"two-fold:":[217],"First,":[218],"design":[220,255,276,319,360],"induce":[226,280],"supervision":[229],"at":[230],"levels.":[232],"And":[233],"second,":[234],"since":[235],"novel,":[239],"evaluations":[243],"show":[245],"effectiveness.":[247],"Therefore,":[248],"every":[250],"model":[251,278,308,321,362],"develop,":[253],"and/or":[256],"gather":[257],"set":[259],"evaluate":[263,306],"interpret":[265],"our":[269,301],"first":[273],"chapter,":[274,317,348],"morpheme-like":[281],"sequence":[285,328],"characters.":[287],"We":[288,303],"adapt":[289],"method":[291],"object":[293],"discovery":[294],"vision,":[296],"called":[297],"Slot":[298],"Attention":[299],"purpose.":[302],"propose":[304],"introducing":[310],"bi-directional":[311],"probing":[312],"evaluation.":[313],"second":[316],"induces":[323,364],"word-like":[324],"integrating":[332],"non-parametric":[333],"variational":[334],"information":[335],"bottleneck":[336],"last":[339],"transformer":[343],"encoder.":[344],"next":[347],"move":[350],"multi-modal":[353],"domain":[354],"starting":[356],"phrases":[365],"image":[367],"captions":[368],"aligning":[370],"objects":[374],"image.":[377],"Lastly,":[378],"explore":[380],"task-driven":[382],"entities.":[386]},"counts_by_year":[],"updated_date":"2026-03-20T23:20:44.827607","created_date":"2025-10-10T00:00:00"}