{"id":"https://openalex.org/W4415036744","doi":"https://doi.org/10.48550/arxiv.2505.21058","title":"Disentangling Locality and Entropy in Ranking Distillation","display_name":"Disentangling Locality and Entropy in Ranking Distillation","publication_year":2025,"publication_date":"2025-05-27","ids":{"openalex":"https://openalex.org/W4415036744","doi":"https://doi.org/10.48550/arxiv.2505.21058"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2505.21058","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.21058","pdf_url":"https://arxiv.org/pdf/2505.21058","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2505.21058","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075770550","display_name":"Andrew Parry","orcid":"https://orcid.org/0000-0001-5446-8328","observed_orcids":["https://orcid.org/0000-0001-5446-8328"]},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Parry, Andrew","raw_affiliation_strings":["School of Computing Science , University of Glasgow"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing Science , University of Glasgow","institution_ids":["https://openalex.org/I7882870"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082339849","display_name":"Debasis Ganguly","orcid":"https://orcid.org/0000-0003-0050-7138","observed_orcids":["https://orcid.org/0000-0003-0050-7138"]},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"Ganguly, Debasis","raw_affiliation_strings":["School of Computing Science , University of Glasgow"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing Science , University of Glasgow","institution_ids":["https://openalex.org/I7882870"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5014199889","display_name":"Sean MacAvaney","orcid":"https://orcid.org/0000-0002-8914-2659","observed_orcids":["https://orcid.org/0000-0002-8914-2659"]},"institutions":[{"id":"https://openalex.org/I7882870","display_name":"University of Glasgow","ror":"https://ror.org/00vtgdb53","country_code":"GB","type":"education","lineage":["https://openalex.org/I7882870"]}],"countries":["GB"],"is_corresponding":false,"raw_author_name":"MacAvaney, Sean","raw_affiliation_strings":["School of Computing Science , University of Glasgow"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Computing Science , University of Glasgow","institution_ids":["https://openalex.org/I7882870"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I7882870"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":1,"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.9351000189781189,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10286","display_name":"Information Retrieval and Search Behavior","score":0.9351000189781189,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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":0.05640000104904175,"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/T10203","display_name":"Recommender Systems and Techniques","score":0.0027000000700354576,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/neural-ranking","display_name":"neural ranking","score":0.8069999814033508},{"id":"https://openalex.org/keywords/knowledge-distillation","display_name":"knowledge distillation","score":0.949999988079071},{"id":"https://openalex.org/keywords/hard-negative-sampling","display_name":"hard negative sampling","score":0.9010000228881836},{"id":"https://openalex.org/keywords/dataset-selection","display_name":"data selection","score":0.5180000066757202},{"id":"https://openalex.org/keywords/teacher-model","display_name":"teacher model","score":0.46799999475479126},{"id":"https://openalex.org/keywords/data-augmentation","display_name":"data augmentation","score":0.8830000162124634},{"id":"https://openalex.org/keywords/semantic-search","display_name":"semantic search","score":0.7020000219345093},{"id":"https://openalex.org/keywords/training-dynamics","display_name":"training dynamics","score":0.5360000133514404},{"id":"https://openalex.org/keywords/example-selection","display_name":"example selection","score":0.47699999809265137},{"id":"https://openalex.org/keywords/contrastive-learning","display_name":"contrastive learning","score":0.4779999852180481},{"id":"https://openalex.org/keywords/sentence-transformers","display_name":"Sentence Transformers","score":0.47999998927116394}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6941999793052673},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.6468999981880188},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.613099992275238},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5823000073432922},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.550599992275238},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.5056999921798706},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5034999847412109},{"id":"https://openalex.org/C106301342","wikidata":"https://www.wikidata.org/wiki/Q4117933","display_name":"Entropy (arrow of time)","level":2,"score":0.49900001287460327},{"id":"https://openalex.org/C2779808786","wikidata":"https://www.wikidata.org/wiki/Q6664603","display_name":"Locality","level":2,"score":0.3930000066757202},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3797999918460846},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.37220001220703125},{"id":"https://openalex.org/C37381756","wikidata":"https://www.wikidata.org/wiki/Q20203288","display_name":"Representativeness heuristic","level":2,"score":0.34360000491142273},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.3434999883174896},{"id":"https://openalex.org/C124975894","wikidata":"https://www.wikidata.org/wiki/Q7293290","display_name":"Ranking SVM","level":3,"score":0.3352000117301941},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3138999938964844},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.31130000948905945},{"id":"https://openalex.org/C75917345","wikidata":"https://www.wikidata.org/wiki/Q2725298","display_name":"Sampling bias","level":3,"score":0.28619998693466187},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.28200000524520874},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.27219998836517334},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2694000005722046},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.2639000117778778},{"id":"https://openalex.org/C86037889","wikidata":"https://www.wikidata.org/wiki/Q4330127","display_name":"Learning to rank","level":3,"score":0.2621999979019165},{"id":"https://openalex.org/C93959086","wikidata":"https://www.wikidata.org/wiki/Q6888345","display_name":"Model selection","level":2,"score":0.2522999942302704}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2505.21058","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.21058","pdf_url":"https://arxiv.org/pdf/2505.21058","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2505.21058","doi":"https://doi.org/10.48550/arxiv.2505.21058","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2505.21058","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2505.21058","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2505.21058","pdf_url":"https://arxiv.org/pdf/2505.21058","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4415036744.pdf","grobid_xml":"https://content.openalex.org/works/W4415036744.grobid-xml"},"referenced_works_count":27,"referenced_works":["https://openalex.org/W1821462560","https://openalex.org/W1971361630","https://openalex.org/W2069870183","https://openalex.org/W2130700086","https://openalex.org/W2610935556","https://openalex.org/W2909544278","https://openalex.org/W2963341956","https://openalex.org/W2970641574","https://openalex.org/W2980282514","https://openalex.org/W2995615678","https://openalex.org/W3015883388","https://openalex.org/W3100107515","https://openalex.org/W3155895380","https://openalex.org/W3168875417","https://openalex.org/W3208821253","https://openalex.org/W4233413206","https://openalex.org/W4284674986","https://openalex.org/W4284699573","https://openalex.org/W4287645694","https://openalex.org/W4387968184","https://openalex.org/W4388444360","https://openalex.org/W4388955489","https://openalex.org/W4389469970","https://openalex.org/W4389523765","https://openalex.org/W4394971440","https://openalex.org/W4396787170","https://openalex.org/W4412394816"],"related_works":["https://openalex.org/W7133194201","https://openalex.org/W4404355224","https://openalex.org/W4401330148","https://openalex.org/W4415348295","https://openalex.org/W4286950596","https://openalex.org/W2808847742","https://openalex.org/W2951780535","https://openalex.org/W4309045033","https://openalex.org/W4319793489","https://openalex.org/W4287645694"],"abstract_inverted_index":{"The":[0],"training":[1,117,245],"process":[2],"of":[3,100,116,125,141,151,209,220],"ranking":[4,20,48,153,156,171],"models":[5,74,109,221],"involves":[6],"two":[7],"key":[8],"data":[9,79,163,205],"selection":[10,147],"decisions:":[11],"a":[12,16,30,42,51,122,170],"sampling":[13,33,87,126,184],"strategy,":[14],"and":[15,41,63,88,111,127,135,148,179,211,216,241],"labeling":[17,44,89],"strategy.":[18],"Modern":[19],"systems,":[21],"especially":[22],"those":[23],"for":[24,65],"performing":[25],"semantic":[26],"search,":[27],"typically":[28],"use":[29],"``hard":[31],"negative''":[32],"strategy":[34,45],"to":[35,46,105,229],"identify":[36],"challenging":[37],"items":[38],"using":[39],"heuristics":[40],"distillation":[43,128],"transfer":[47],"\"knowledge\"":[49],"from":[50,70],"more":[52,231],"capable":[53],"model.":[54,172],"In":[55],"practice,":[56],"these":[57],"approaches":[58,234],"have":[59,91],"grown":[60],"increasingly":[61],"expensive":[62],"complex,":[64],"instance,":[66],"popular":[67],"pretrained":[68],"rankers":[69],"SentenceTransformers":[71],"involve":[72],"12":[73],"in":[75,130,161,169,191,207,222],"an":[76],"ensemble":[77],"with":[78],"provenance":[80],"hampering":[81],"reproducibility.":[82],"Despite":[83],"their":[84],"complexity,":[85],"modern":[86],"strategies":[90],"not":[92],"been":[93],"fully":[94],"ablated,":[95],"leaving":[96],"the":[97,114,138,149,217],"underlying":[98],"source":[99],"effectiveness":[101,215],"gains":[102],"unclear.":[103],"Thus,":[104],"better":[106,250],"understand":[107,244],"why":[108],"improve":[110,167],"potentially":[112],"reduce":[113,236],"expense":[115],"effective":[118,190],"models,":[119],"we":[120,227],"conduct":[121],"broad":[123],"ablation":[124],"processes":[129,185],"neural":[131],"ranking.":[132,223],"We":[133,201],"frame":[134],"theoretically":[136],"derive":[137],"orthogonal":[139],"nature":[140],"model":[142,157],"geometry":[143],"affected":[144],"by":[145],"example":[146],"effect":[150],"teacher":[152],"entropy":[154],"on":[155,176,238],"optimization,":[158],"establishing":[159],"conditions":[160],"which":[162,249],"augmentation":[164],"can":[165,213],"effectively":[166],"bias":[168],"Empirically,":[173],"our":[174],"investigation":[175],"established":[177],"benchmarks":[178],"common":[180],"architectures":[181],"shows":[182],"that":[183,186,235],"were":[187],"once":[188],"highly":[189],"contrastive":[192,239],"objectives":[193],"may":[194],"be":[195],"spurious":[196],"or":[197],"harmful":[198],"under":[199,247],"distillation.":[200],"further":[202],"investigate":[203],"how":[204],"augmentation,":[206],"terms":[208],"inputs":[210],"targets,":[212],"affect":[214],"intrinsic":[218],"behavior":[219],"Through":[224],"this":[225],"work,":[226],"aim":[228],"encourage":[230],"computationally":[232],"efficient":[233],"focus":[237],"pairs":[240],"instead":[242],"directly":[243],"dynamics":[246],"rankings,":[248],"represent":[251],"real-world":[252],"settings.":[253]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}