{"id":"https://openalex.org/W7154941062","doi":"https://doi.org/10.48550/arxiv.2604.16197","title":"Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation","display_name":"Sketching the Readout of Large Language Models for Scalable Data Attribution and Valuation","publication_year":2026,"publication_date":"2026-04-17","ids":{"openalex":"https://openalex.org/W7154941062","doi":"https://doi.org/10.48550/arxiv.2604.16197"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.16197","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.16197","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2604.16197","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134032041","display_name":"Yide Ran","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ran, Yide","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134040537","display_name":"Jianwen Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Jianwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134033927","display_name":"Minghui Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Minghui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134092491","display_name":"Wenjin Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Wenjin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134072298","display_name":"Denghui Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Denghui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5017970736","display_name":"Li Chuan","orcid":"https://orcid.org/0000-0002-7773-6310"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Chuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134076015","display_name":"Zhaozhuo Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Zhaozhuo","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/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.3450999855995178,"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"}},"topics":[{"id":"https://openalex.org/T11636","display_name":"Artificial Intelligence in Healthcare and Education","score":0.3450999855995178,"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":0.19900000095367432,"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/T13702","display_name":"Machine Learning in Healthcare","score":0.07559999823570251,"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/scalability","display_name":"Scalability","score":0.6991000175476074},{"id":"https://openalex.org/keywords/search-engine-indexing","display_name":"Search engine indexing","score":0.6848000288009644},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5461000204086304},{"id":"https://openalex.org/keywords/residual","display_name":"Residual","score":0.4106999933719635},{"id":"https://openalex.org/keywords/external-data-representation","display_name":"External Data Representation","score":0.3709000051021576},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.3702999949455261},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.36579999327659607},{"id":"https://openalex.org/keywords/channel","display_name":"Channel (broadcasting)","score":0.3601999878883362},{"id":"https://openalex.org/keywords/valuation","display_name":"Valuation (finance)","score":0.35120001435279846}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7939000129699707},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6991000175476074},{"id":"https://openalex.org/C75165309","wikidata":"https://www.wikidata.org/wiki/Q2258979","display_name":"Search engine indexing","level":2,"score":0.6848000288009644},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5461000204086304},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.44589999318122864},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.42149999737739563},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.4106999933719635},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3970000147819519},{"id":"https://openalex.org/C116409475","wikidata":"https://www.wikidata.org/wiki/Q1385056","display_name":"External Data Representation","level":2,"score":0.3709000051021576},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.3702999949455261},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.36579999327659607},{"id":"https://openalex.org/C127162648","wikidata":"https://www.wikidata.org/wiki/Q16858953","display_name":"Channel (broadcasting)","level":2,"score":0.3601999878883362},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3589000105857849},{"id":"https://openalex.org/C186027771","wikidata":"https://www.wikidata.org/wiki/Q4008379","display_name":"Valuation (finance)","level":2,"score":0.35120001435279846},{"id":"https://openalex.org/C199163554","wikidata":"https://www.wikidata.org/wiki/Q1681619","display_name":"Univariate","level":3,"score":0.32679998874664307},{"id":"https://openalex.org/C43711488","wikidata":"https://www.wikidata.org/wiki/Q7534783","display_name":"Skew","level":2,"score":0.31470000743865967},{"id":"https://openalex.org/C162319229","wikidata":"https://www.wikidata.org/wiki/Q175263","display_name":"Data structure","level":2,"score":0.3059000074863434},{"id":"https://openalex.org/C75684735","wikidata":"https://www.wikidata.org/wiki/Q858810","display_name":"Big data","level":2,"score":0.305400013923645},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.30379998683929443},{"id":"https://openalex.org/C168065819","wikidata":"https://www.wikidata.org/wiki/Q845566","display_name":"Debugging","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C134537474","wikidata":"https://www.wikidata.org/wiki/Q17144832","display_name":"Naturalness","level":2,"score":0.30090001225471497},{"id":"https://openalex.org/C2781045450","wikidata":"https://www.wikidata.org/wiki/Q254569","display_name":"Backdoor","level":2,"score":0.30000001192092896},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.2865000069141388},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2842000126838684},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.2694999873638153},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.2574999928474426},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25600001215934753},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.25200000405311584},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.2508000135421753}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.16197","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.16197","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2604.16197","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.16197","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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/16","score":0.684573769569397,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Data":[0],"attribution":[1],"and":[2,55,75,94,117,132,144,164,186,212,218],"valuation":[3],"are":[4],"critical":[5],"for":[6,10,161,215],"understanding":[7],"data-model":[8],"synergy":[9],"Large":[11],"Language":[12],"Models":[13],"(LLMs),":[14],"yet":[15],"existing":[16],"gradient-based":[17,139],"methods":[18],"suffer":[19],"from":[20],"scalability":[21],"challenges":[22],"on":[23,33,64,151,176,200],"LLMs.":[24],"Inspired":[25],"by":[26,125],"human":[27],"cognition,":[28],"where":[29,71,138],"decision":[30],"making":[31],"relies":[32],"a":[34,79,85,89,95,193,210],"focused":[35],"readout":[36],"of":[37,53],"relevant":[38],"memories":[39],"rather":[40],"than":[41],"replaying":[42],"all":[43],"pathways,":[44],"we":[45],"introduce":[46],"RISE":[47,62,121,150,175,208],"(Readout":[48],"Influence":[49],"Sketching":[50],"Estimator).":[51],"Instead":[52],"computing":[54],"indexing":[56],"gradients":[57],"across":[58],"the":[59,68,76,114],"entire":[60],"LLM,":[61,137],"focuses":[63],"influence":[65,72,216],"hotspots":[66],"at":[67],"output":[69],"layer,":[70],"signals":[73],"concentrate,":[74],"gradient":[77],"admits":[78],"decomposed":[80],"outer-product":[81],"form.":[82],"This":[83],"enables":[84],"dual-channel":[86],"representation":[87],"combining":[88],"lexical":[90],"residual":[91],"channel":[92,98],"(RH)":[93],"semantic":[96],"projected-error":[97],"(GH).":[99],"Applying":[100],"CountSketch":[101],"projections":[102],"to":[103,127,130,134],"these":[104],"channels":[105],"achieves":[106],"strong":[107],"compression":[108],"while":[109],"maintaining":[110],"accurate":[111],"attribution.":[112],"Across":[113],"OLMo":[115],"(1B-32B)":[116],"Pythia":[118],"(14M-6.9B)":[119],"families,":[120],"reduces":[122],"index":[123],"storage":[124],"up":[126],"112$\\times$":[128],"compared":[129],"RapidIn":[131,143],"scales":[133],"32B":[135],"parameters":[136],"baselines":[140],"such":[141],"as":[142],"ZO-Inf":[145],"become":[146],"memory-infeasible.":[147],"We":[148,173],"evaluate":[149],"two":[152],"paradigms:":[153],"(1)":[154],"retrospective":[155],"attribution,":[156],"retrieving":[157],"influential":[158],"training":[159],"examples":[160],"specific":[162],"predictions,":[163],"(2)":[165],"prospective":[166],"valuation,":[167],"scoring":[168],"candidate":[169],"data":[170,181,190,202],"utility":[171],"zero-shot.":[172],"validate":[174],"three":[177],"tasks:":[178],"Howdy":[179],"backdoor":[180],"detection,":[182],"Finance-Medical":[183],"domain":[184],"separation,":[185],"Brain":[187,195],"Rot":[188,196],"high-quality":[189],"selection.":[191],"In":[192],"closed-loop":[194],"study,":[197],"continued":[198],"pretraining":[199],"RISE-selected":[201],"yields":[203],"consistent":[204],"downstream":[205],"improvements.":[206],"Overall,":[207],"provides":[209],"practical":[211],"scalable":[213],"primitive":[214],"analysis":[217],"training-data":[219],"selection":[220],"in":[221],"modern":[222],"large":[223],"language":[224],"models.":[225]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-21T00:00:00"}