{"id":"https://openalex.org/W7161023433","doi":"https://doi.org/10.48550/arxiv.2605.11774","title":"From Token to Token Pair: Efficient Prompt Compression for Large Language Models in Clinical Prediction","display_name":"From Token to Token Pair: Efficient Prompt Compression for Large Language Models in Clinical Prediction","publication_year":2026,"publication_date":"2026-05-12","ids":{"openalex":"https://openalex.org/W7161023433","doi":"https://doi.org/10.48550/arxiv.2605.11774"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.11774","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11774","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,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.11774","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136063884","display_name":"Mingcheng Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Mingcheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136058583","display_name":"Zhiyao Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Zhiyao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136074666","display_name":"Yu Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136058381","display_name":"Tingting Zhu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Tingting","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.9538999795913696,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.9538999795913696,"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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.01810000091791153,"subfield":{"id":"https://openalex.org/subfields/2718","display_name":"Health Informatics"},"field":{"id":"https://openalex.org/fields/27","display_name":"Medicine"},"domain":{"id":"https://openalex.org/domains/4","display_name":"Health Sciences"}},{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.007000000216066837,"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/security-token","display_name":"Security token","score":0.8396000266075134},{"id":"https://openalex.org/keywords/lossless-compression","display_name":"Lossless compression","score":0.692799985408783},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5584999918937683},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.5105000138282776},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5058000087738037},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.4973999857902527},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.454800009727478},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.4034000039100647}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.8396000266075134},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7699999809265137},{"id":"https://openalex.org/C81081738","wikidata":"https://www.wikidata.org/wiki/Q55542","display_name":"Lossless compression","level":3,"score":0.692799985408783},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5584999918937683},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.5105000138282776},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5058000087738037},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.4973999857902527},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.454800009727478},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.4034000039100647},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3612000048160553},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3601999878883362},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.35850000381469727},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3391999900341034},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.31150001287460327},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.30799999833106995},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.29280000925064087},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.2856000065803528},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2840999960899353},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.28110000491142273},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.28049999475479126},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.27959999442100525},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2750999927520752},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2728999853134155},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.27239999175071716},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.2718000113964081},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2619999945163727}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.11774","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11774","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,"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":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.11774","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.11774","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,"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":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/3","display_name":"Good health and well-being","score":0.6197561025619507}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"By":[0],"processing":[1],"electronic":[2],"health":[3],"records":[4],"(EHRs)":[5],"as":[6,22],"natural":[7],"language":[8,11],"sequences,":[9],"large":[10],"models":[12],"(LLMs)":[13],"have":[14],"shown":[15],"potential":[16],"in":[17,41],"clinical":[18,69,150,185],"prediction":[19,24,186],"tasks":[20],"such":[21],"mortality":[23],"and":[25,45,163,176,183,197,201,204],"phenotyping.":[26],"However,":[27],"longitudinal":[28],"or":[29,56,66,81,170],"highly":[30],"frequent":[31],"EHRs":[32],"often":[33],"yield":[34],"excessively":[35],"long":[36],"token":[37,76,106,157],"sequences":[38,77],"that":[39,94,153],"result":[40],"high":[42],"computational":[43,117],"costs":[44],"even":[46,171],"reduced":[47],"performance.":[48],"Existing":[49],"solutions":[50],"either":[51],"add":[52],"modules":[53],"for":[54,98,148],"compression":[55,74,113],"remove":[57],"less":[58],"important":[59],"tokens,":[60,110],"which":[61],"introduce":[62],"additional":[63,79],"inference":[64,164],"latency":[65,165],"risk":[67],"losing":[68],"information.":[70],"To":[71],"achieve":[72],"lossless":[73,112],"of":[75,83,127,132,135],"without":[78],"cost":[80],"loss":[82],"performance,":[84],"we":[85],"propose":[86],"Medical":[87],"Token-Pair":[88],"Encoding":[89],"(MedTPE),":[90],"a":[91,120],"layered":[92],"method":[93],"extends":[95],"standard":[96],"tokenisation":[97],"EHR":[99],"sequences.":[100],"MedTPE":[101,154,189],"merges":[102],"frequently":[103],"co-occurring":[104],"medical":[105],"pairs":[107],"into":[108],"composite":[109],"providing":[111],"while":[114,168],"preserving":[115],"the":[116,125,128,136],"complexity":[118],"through":[119],"dependency-aware":[121],"replacement":[122],"strategy.":[123],"Only":[124],"embeddings":[126],"newly":[129],"introduced":[130],"tokens":[131],"merely":[133],"0.5-1.0%":[134],"LLM's":[137],"parameters":[138],"are":[139],"fine-tuned":[140],"via":[141],"self-supervised":[142],"learning.":[143],"Experiments":[144],"on":[145],"real-world":[146],"datasets":[147],"two":[149],"scenarios":[151],"demonstrate":[152],"reduces":[155],"input":[156,194],"length":[158],"by":[159,166],"up":[160],"to":[161,199],"31%":[162],"34-63%,":[167],"maintaining":[169],"improving":[172],"both":[173],"predictive":[174],"performance":[175],"output":[177],"format":[178],"compliance":[179],"across":[180,192],"multiple":[181],"LLMs":[182],"four":[184],"tasks.":[187],"Furthermore,":[188],"demonstrates":[190],"robustness":[191],"different":[193,205],"context":[195],"lengths":[196],"generalisability":[198],"scientific":[200],"financial":[202],"domains":[203],"languages.":[206]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-05-14T00:00:00"}