{"id":"https://openalex.org/W4385568477","doi":"https://doi.org/10.1145/3580305.3599402","title":"LEA: Improving Sentence Similarity Robustness to Typos Using Lexical Attention Bias","display_name":"LEA: Improving Sentence Similarity Robustness to Typos Using Lexical Attention Bias","publication_year":2023,"publication_date":"2023-08-04","ids":{"openalex":"https://openalex.org/W4385568477","doi":"https://doi.org/10.1145/3580305.3599402"},"language":"en","primary_location":{"id":"doi:10.1145/3580305.3599402","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3580305.3599402","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5045149275","display_name":"Mario Almagro","orcid":"https://orcid.org/0000-0003-4339-2959"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Mario Almagro","raw_affiliation_strings":["NielsenIQ Innovation, Madrid, Spain"],"affiliations":[{"raw_affiliation_string":"NielsenIQ Innovation, Madrid, Spain","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073663561","display_name":"Emilio Almaz\u00e1n","orcid":"https://orcid.org/0009-0008-1583-3749"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Emilio Almaz\u00e1n","raw_affiliation_strings":["NielsenIQ Innovation, Madrid, Spain"],"affiliations":[{"raw_affiliation_string":"NielsenIQ Innovation, Madrid, Spain","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028186506","display_name":"Diego Ortego","orcid":"https://orcid.org/0000-0002-1011-3610"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Diego Ortego","raw_affiliation_strings":["NielsenIQ Innovation, Madrid, Spain"],"affiliations":[{"raw_affiliation_string":"NielsenIQ Innovation, Madrid, Spain","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101416378","display_name":"David Jim\u00e9nez","orcid":"https://orcid.org/0000-0002-7194-5541"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"David Jim\u00e9nez","raw_affiliation_strings":["NielsenIQ Innovation, Madrid, Spain"],"affiliations":[{"raw_affiliation_string":"NielsenIQ Innovation, Madrid, Spain","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":["https://openalex.org/A5045149275"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.049,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.8108323,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"36","last_page":"46"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.9998000264167786,"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":0.9998000264167786,"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":0.9997000098228455,"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/T10664","display_name":"Sentiment Analysis and Opinion Mining","score":0.9973000288009644,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8279885053634644},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6795731782913208},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6238849759101868},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.6167536973953247},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.5970885753631592},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5900633335113525},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5544947385787964},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5434496998786926},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.4163576662540436},{"id":"https://openalex.org/keywords/lexical-analysis","display_name":"Lexical analysis","score":0.41518503427505493},{"id":"https://openalex.org/keywords/speech-recognition","display_name":"Speech recognition","score":0.34440985321998596}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8279885053634644},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6795731782913208},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6238849759101868},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.6167536973953247},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.5970885753631592},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5900633335113525},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5544947385787964},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5434496998786926},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.4163576662540436},{"id":"https://openalex.org/C176982825","wikidata":"https://www.wikidata.org/wiki/Q835922","display_name":"Lexical analysis","level":2,"score":0.41518503427505493},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.34440985321998596},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.0},{"id":"https://openalex.org/C55493867","wikidata":"https://www.wikidata.org/wiki/Q7094","display_name":"Biochemistry","level":1,"score":0.0},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3580305.3599402","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3580305.3599402","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":9,"referenced_works":["https://openalex.org/W569478347","https://openalex.org/W2525579820","https://openalex.org/W2592549418","https://openalex.org/W2931479232","https://openalex.org/W2964110616","https://openalex.org/W2984051011","https://openalex.org/W3175475869","https://openalex.org/W4221163653","https://openalex.org/W4224980447"],"related_works":["https://openalex.org/W2151197196","https://openalex.org/W2949968076","https://openalex.org/W2935811960","https://openalex.org/W2969407538","https://openalex.org/W2917049097","https://openalex.org/W4298195702","https://openalex.org/W3045818198","https://openalex.org/W3034697300","https://openalex.org/W4287760117","https://openalex.org/W3093768914"],"abstract_inverted_index":{"Textual":[0],"noise,":[1,169,202],"such":[2],"as":[3],"typos":[4,230],"or":[5,39],"abbreviations,":[6],"is":[7,23,227],"a":[8,30,122],"well-known":[9],"issue":[10,71],"that":[11,21,86,128,154,193,225],"penalizes":[12],"vanilla":[13],"Transformers":[14],"for":[15,27,93,189,219],"most":[16],"downstream":[17],"tasks.":[18],"We":[19,152,211],"show":[20,192],"this":[22,110],"also":[24,212],"the":[25,49,55,58,62,69,90,102,145,155,199,207,253,256],"case":[26],"sentence":[28],"similarity,":[29],"fundamental":[31],"task":[32],"in":[33,54,134,171,185,216,231,247,263],"multiple":[34],"domains,":[35],"e.g.":[36],"matching,":[37],"retrieval":[38],"paraphrasing.":[40],"Sentence":[41],"similarity":[42],"can":[43],"be":[44],"approached":[45],"using":[46,138,180],"cross-encoders,":[47],"where":[48],"two":[50,217],"sentences":[51,235],"are":[52,87],"concatenated":[53],"input":[56],"allowing":[57],"model":[59],"to":[60,89,114,163,229],"exploit":[61],"inter-relations":[63],"between":[64,132],"them.":[65],"Previous":[66],"works":[67],"addressing":[68],"noise":[70,117],"mainly":[72],"rely":[73],"on":[74,206],"data":[75],"augmentation":[76],"strategies,":[77],"showing":[78,224],"improved":[79,150],"robustness":[80],"when":[81],"dealing":[82,265],"with":[83,121,167,173,233,266],"corrupted":[84],"samples":[85],"similar":[88],"ones":[91],"used":[92],"training.":[94],"However,":[95],"all":[96],"these":[97],"methods":[98],"still":[99],"suffer":[100],"from":[101],"token":[103],"distribution":[104],"shift":[105,147],"induced":[106],"by":[107,118,159],"typos.":[108,267],"In":[109],"work,":[111],"we":[112,241],"propose":[113],"tackle":[115,164],"textual":[116,168,220],"equipping":[119],"cross-encoders":[120,162,264],"novel":[123],"LExical-aware":[124],"Attention":[125],"module":[126],"(LEA)":[127],"incorporates":[129],"lexical":[130],"similarities":[131],"words":[133],"both":[135],"sentences.":[136],"By":[137],"raw":[139],"text":[140],"similarities,":[141],"our":[142,214,248],"approach":[143,215],"avoids":[144],"tokenization":[146],"problem":[148],"obtaining":[149],"robustness.":[151],"demonstrate":[153],"attention":[156],"bias":[157],"introduced":[158],"LEA":[160,194,226],"helps":[161],"complex":[165],"scenarios":[166],"specially":[170],"domains":[172,232],"short-text":[174],"descriptions":[175],"and":[176,222,236,259],"limited":[177],"context.":[178,239],"Experiments":[179],"three":[181],"popular":[182],"Transformer":[183],"encoders":[184],"five":[186],"e-commerce":[187],"datasets":[188,218],"product":[190],"matching":[191],"consistently":[195],"boosts":[196],"performance":[197],"under":[198],"presence":[200],"of":[201,255],"while":[203],"remaining":[204],"competitive":[205],"original":[208],"(clean)":[209],"splits.":[210],"evaluate":[213],"entailment":[221],"paraphrasing":[223],"robust":[228],"longer":[234],"more":[237],"natural":[238],"Additionally,":[240],"thoroughly":[242],"analyze":[243],"several":[244],"design":[245],"choices":[246],"approach,":[249],"providing":[250],"insights":[251],"about":[252],"impact":[254],"decisions":[257],"made":[258],"fostering":[260],"future":[261],"research":[262]},"counts_by_year":[{"year":2026,"cited_by_count":2},{"year":2025,"cited_by_count":2},{"year":2024,"cited_by_count":1},{"year":2023,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}