{"id":"https://openalex.org/W4412889374","doi":"https://doi.org/10.18653/v1/2025.acl-srw.51","title":"Low-Perplexity LLM-Generated Sequences and Where To Find Them","display_name":"Low-Perplexity LLM-Generated Sequences and Where To Find Them","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412889374","doi":"https://doi.org/10.18653/v1/2025.acl-srw.51"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.acl-srw.51","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-srw.51","pdf_url":"https://aclanthology.org/2025.acl-srw.51.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.acl-srw.51.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5119181738","display_name":"Arthur Wuhrmann","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Arthur Wuhrmann","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5067556590","display_name":"Andrei Kucharavy","orcid":"https://orcid.org/0000-0003-0429-8644"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Andrei Kucharavy","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5119181740","display_name":"Anastasiia Kucherenko","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anastasiia Kucherenko","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5119181738"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.09174713,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"774","last_page":"783"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11130","display_name":"Coding theory and cryptography","score":0.9930999875068665,"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/T11130","display_name":"Coding theory and cryptography","score":0.9930999875068665,"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/perplexity","display_name":"Perplexity","score":0.9345880150794983},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6074826121330261},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.28502243757247925},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.11970958113670349}],"concepts":[{"id":"https://openalex.org/C100279451","wikidata":"https://www.wikidata.org/wiki/Q372193","display_name":"Perplexity","level":3,"score":0.9345880150794983},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6074826121330261},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.28502243757247925},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.11970958113670349}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.18653/v1/2025.acl-srw.51","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-srw.51","pdf_url":"https://aclanthology.org/2025.acl-srw.51.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)","raw_type":"proceedings-article"},{"id":"pmh:oai:hesso.tind.io:17761","is_oa":false,"landing_page_url":"http://arodes.hes-so.ch/record/17761","pdf_url":null,"source":{"id":"https://openalex.org/S4306402432","display_name":"ArODES (HES-SO (https://www.hes-so.ch/))","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I4210088449","host_organization_name":"HES-SO Gen\u00e8ve","host_organization_lineage":["https://openalex.org/I4210088449"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"http://arodes.hes-so.ch/record/17761","raw_type":"Text"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.acl-srw.51","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.acl-srw.51","pdf_url":"https://aclanthology.org/2025.acl-srw.51.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412889374.pdf","grobid_xml":"https://content.openalex.org/works/W4412889374.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2376415519","https://openalex.org/W1601381279","https://openalex.org/W4294769427","https://openalex.org/W4225667838","https://openalex.org/W1895908943","https://openalex.org/W4281893144","https://openalex.org/W2374747083"],"abstract_inverted_index":{"As":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"become":[5],"increasingly":[6],"widespread,":[7],"understanding":[8,115],"how":[9,25,117],"specific":[10],"training":[11,31,70,119],"data":[12,120],"shapes":[13],"their":[14,30,66,122],"outputs":[15],"is":[16],"crucial":[17],"for":[18],"transparency,":[19],"accountability,":[20],"privacy,":[21],"and":[22,28,104,109],"fairness.To":[23],"explore":[24],"LLMs":[26,118],"leverage":[27],"replicate":[29],"data,":[32],"we":[33,72,92],"introduce":[34],"a":[35,75,111],"systematic":[36],"approach":[37],"centered":[38],"on":[39],"analyzing":[40],"lowperplexity":[41],"sequences-high-probability":[42],"text":[43],"spans":[44,81],"generated":[45],"by":[46],"the":[47,69,86,94,102],"model.Our":[48],"pipeline":[49],"reliably":[50],"extracts":[51],"such":[52],"long":[53],"sequences":[54],"across":[55,98],"diverse":[56],"topics":[57],"while":[58],"avoiding":[59],"degeneration,":[60],"then":[61],"traces":[62],"them":[63],"back":[64],"to":[65,85],"sources":[67],"in":[68],"data.Surprisingly,":[71],"find":[73],"that":[74,89],"substantial":[76],"portion":[77],"of":[78,96,106,116],"these":[79],"low-perplexity":[80],"cannot":[82],"be":[83],"mapped":[84],"corpus.For":[87],"those":[88],"do":[90],"match,":[91],"quantify":[93],"distribution":[95],"occurrences":[97],"source":[99],"documents,":[100],"highlighting":[101],"scope":[103],"nature":[105],"verbatim":[107],"recall":[108],"paving":[110],"way":[112],"toward":[113],"better":[114],"impacts":[121],"behavior.":[123]},"counts_by_year":[],"updated_date":"2026-05-05T08:41:31.759640","created_date":"2025-10-10T00:00:00"}