{"id":"https://openalex.org/W7165626180","doi":"https://doi.org/10.48550/arxiv.2606.23591","title":"Quantifying the Agreement Between Data-Influence and Data-Similarity to Understand LLM Behavior","display_name":"Quantifying the Agreement Between Data-Influence and Data-Similarity to Understand LLM Behavior","publication_year":2026,"publication_date":"2026-06-22","ids":{"openalex":"https://openalex.org/W7165626180","doi":"https://doi.org/10.48550/arxiv.2606.23591"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.23591","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23591","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","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.2606.23591","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075097456","display_name":"Christopher J. Anders","orcid":"https://orcid.org/0000-0003-3295-8486"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Anders, Christopher J.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114418433","display_name":"Henrique Da Silva Gameiro","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gameiro, Henrique Da Silva","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075130141","display_name":"Nico Daheim","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Daheim, Nico","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5053565482","display_name":"Mohammad Emtiyaz Khan","orcid":"https://orcid.org/0000-0001-7614-0489"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Khan, Mohammad Emtiyaz","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/T10028","display_name":"Topic Modeling","score":0.24580000340938568,"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/T10028","display_name":"Topic Modeling","score":0.24580000340938568,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.2199999988079071,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.05009999871253967,"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/measure","display_name":"Measure (data warehouse)","score":0.746999979019165},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.6409000158309937},{"id":"https://openalex.org/keywords/ranking","display_name":"Ranking (information retrieval)","score":0.6151000261306763},{"id":"https://openalex.org/keywords/asymmetry","display_name":"Asymmetry","score":0.5964000225067139},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.5791000127792358},{"id":"https://openalex.org/keywords/trace","display_name":"TRACE (psycholinguistics)","score":0.4523000121116638}],"concepts":[{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.746999979019165},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.6409000158309937},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6151000261306763},{"id":"https://openalex.org/C38976095","wikidata":"https://www.wikidata.org/wiki/Q752641","display_name":"Asymmetry","level":2,"score":0.5964000225067139},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.5791000127792358},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5404000282287598},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.4523000121116638},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.37790000438690186},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3472000062465668},{"id":"https://openalex.org/C2776818064","wikidata":"https://www.wikidata.org/wiki/Q829903","display_name":"Agreement","level":2,"score":0.33320000767707825},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3140999972820282},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2856999933719635},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.27639999985694885},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2727999985218048},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.26260000467300415},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.23591","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23591","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.23591","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.23591","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"One":[0],"way":[1,129],"to":[2,7,12,38,85,151,162],"understand":[3],"LLM":[4],"behavior":[5],"is":[6,31,36,100,109,133],"trace":[8],"its":[9],"output":[10,24,57],"back":[11],"the":[13,34,67,72,81,91,94,102,127,149,159,164],"training":[14,82],"data.":[15],"Two":[16],"types":[17],"of":[18,117,138,166],"measures":[19],"are":[20,119],"commonly":[21],"used":[22],"for":[23,49,56],"tracing:":[25],"data-similarity":[26,118],"and":[27,64,69,88,145],"data-influence.":[28],"The":[29,114],"former":[30],"cheaper":[32],"while":[33],"latter":[35],"believed":[37],"be":[39],"more":[40,121],"accurate.":[41],"Even":[42],"though":[43],"many":[44],"works":[45],"have":[46],"compared":[47],"them":[48],"ground-truth":[50],"tasks,":[51],"no":[52],"such":[53],"comparisons":[54],"exist":[55],"tracing.":[58],"Here,":[59],"we":[60],"fill":[61],"this":[62,77],"gap":[63],"precisely":[65],"quantify":[66],"commonalities":[68],"differences":[70],"between":[71,93,112],"two":[73,95,103],"measures.":[74],"We":[75,147],"do":[76],"by":[78,124,157],"first":[79],"ranking":[80],"documents":[83,116],"according":[84],"each":[86],"measure":[87],"then":[89],"computing":[90],"overlap":[92],"rankings.":[96],"Our":[97],"main":[98],"finding":[99],"that":[101],"rankings":[104],"agree":[105],"significantly,":[106],"but":[107],"there":[108],"an":[110],"asymmetry":[111,150],"them:":[113],"top":[115],"assigned":[120],"consistent":[122],"ranks":[123],"data-influence":[125,161],"than":[126],"other":[128],"around.":[130],"This":[131],"result":[132],"valid":[134],"across":[135],"a":[136,153],"range":[137],"experiments":[139],"involving":[140],"OLMo2-1B,":[141],"Qwen3-1.7B,":[142],"LlaMa3.2-1B,":[143],"Gemma3-1B,":[144],"GPT2.":[146],"exploit":[148],"obtain":[152],"favorable":[154],"cost-accuracy":[155],"trade-off":[156],"using":[158],"costly":[160],"refine":[163],"results":[165],"data-similarity.":[167]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-24T00:00:00"}