{"id":"https://openalex.org/W7126209140","doi":"https://doi.org/10.1145/3795137","title":"A Comprehensive Performance Evaluation of LLMs for Data-to-Text Generation and Divergence-Weighted Training","display_name":"A Comprehensive Performance Evaluation of LLMs for Data-to-Text Generation and Divergence-Weighted Training","publication_year":2026,"publication_date":"2026-01-30","ids":{"openalex":"https://openalex.org/W7126209140","doi":"https://doi.org/10.1145/3795137"},"language":"en","primary_location":{"id":"doi:10.1145/3795137","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3795137","pdf_url":null,"source":{"id":"https://openalex.org/S2492086750","display_name":"ACM Transactions on Intelligent Systems and Technology","issn_l":"2157-6904","issn":["2157-6904","2157-6912"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Intelligent Systems and Technology","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124454408","display_name":"Joy Mahapatra","orcid":null},"institutions":[{"id":"https://openalex.org/I6498739","display_name":"Indian Statistical Institute","ror":"https://ror.org/00q2w1j53","country_code":"IN","type":"education","lineage":["https://openalex.org/I6498739"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Joy Mahapatra","raw_affiliation_strings":["Indian Statistical Institute, Kolkata, India"],"raw_orcid":"https://orcid.org/0009-0006-8301-293X","affiliations":[{"raw_affiliation_string":"Indian Statistical Institute, Kolkata, India","institution_ids":["https://openalex.org/I6498739"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5024170958","display_name":"Utpal Garain","orcid":"https://orcid.org/0000-0001-7207-5018"},"institutions":[{"id":"https://openalex.org/I6498739","display_name":"Indian Statistical Institute","ror":"https://ror.org/00q2w1j53","country_code":"IN","type":"education","lineage":["https://openalex.org/I6498739"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Utpal Garain","raw_affiliation_strings":["Indian Statistical Institute, Kolkata, India"],"raw_orcid":"https://orcid.org/0000-0001-7207-5018","affiliations":[{"raw_affiliation_string":"Indian Statistical Institute, Kolkata, India","institution_ids":["https://openalex.org/I6498739"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I6498739"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.10827795,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"17","issue":"2","first_page":"1","last_page":"25"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13629","display_name":"Text Readability and Simplification","score":0.2705000042915344,"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/T13629","display_name":"Text Readability and Simplification","score":0.2705000042915344,"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/T10028","display_name":"Topic Modeling","score":0.1736000031232834,"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/T13910","display_name":"Computational and Text Analysis Methods","score":0.07930000126361847,"subfield":{"id":"https://openalex.org/subfields/3300","display_name":"General Social Sciences"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/readability","display_name":"Readability","score":0.820900022983551},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.6593000292778015},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5153999924659729},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.46790000796318054},{"id":"https://openalex.org/keywords/natural-language-generation","display_name":"Natural language generation","score":0.4465999901294708},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.35420000553131104},{"id":"https://openalex.org/keywords/training-set","display_name":"Training set","score":0.31189998984336853}],"concepts":[{"id":"https://openalex.org/C2778143727","wikidata":"https://www.wikidata.org/wiki/Q1820650","display_name":"Readability","level":2,"score":0.820900022983551},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7997000217437744},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6593000292778015},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5153999924659729},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.46790000796318054},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4537999927997589},{"id":"https://openalex.org/C2776187449","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Natural language generation","level":3,"score":0.4465999901294708},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.41620001196861267},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.36880001425743103},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.35420000553131104},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.31189998984336853},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.29660001397132874},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.2921000123023987},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.27489998936653137},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.2741999924182892},{"id":"https://openalex.org/C3018395757","wikidata":"https://www.wikidata.org/wiki/Q1379672","display_name":"Evaluation methods","level":2,"score":0.26350000500679016},{"id":"https://openalex.org/C134537474","wikidata":"https://www.wikidata.org/wiki/Q17144832","display_name":"Naturalness","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.2606000006198883},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.2533999979496002},{"id":"https://openalex.org/C135510737","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance indicator","level":2,"score":0.2531999945640564}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3795137","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3795137","pdf_url":null,"source":{"id":"https://openalex.org/S2492086750","display_name":"ACM Transactions on Intelligent Systems and Technology","issn_l":"2157-6904","issn":["2157-6904","2157-6912"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Intelligent Systems and Technology","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"score":0.8426103591918945,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":34,"referenced_works":["https://openalex.org/W179875071","https://openalex.org/W2101105183","https://openalex.org/W2604799547","https://openalex.org/W2786660442","https://openalex.org/W2789028076","https://openalex.org/W2798935874","https://openalex.org/W2806532810","https://openalex.org/W2952523122","https://openalex.org/W2962729880","https://openalex.org/W2963532001","https://openalex.org/W2963912046","https://openalex.org/W2970785793","https://openalex.org/W2979826702","https://openalex.org/W2990153872","https://openalex.org/W2995589434","https://openalex.org/W3021150969","https://openalex.org/W3034383590","https://openalex.org/W3034999214","https://openalex.org/W3103317667","https://openalex.org/W3106234277","https://openalex.org/W3166398748","https://openalex.org/W3170806096","https://openalex.org/W3174770825","https://openalex.org/W3201174429","https://openalex.org/W4225593490","https://openalex.org/W4309674289","https://openalex.org/W4379792729","https://openalex.org/W4385565070","https://openalex.org/W4385756501","https://openalex.org/W4389009580","https://openalex.org/W4389518784","https://openalex.org/W4392445491","https://openalex.org/W4392608007","https://openalex.org/W4393156784"],"related_works":[],"abstract_inverted_index":{"Data-to-Text":[0],"Generation":[1],"(D2T)":[2],"aims":[3],"to":[4],"transform":[5],"semi-structured":[6],"data":[7],"(such":[8],"as":[9,30,149,194],"tables":[10],"and":[11,52,57,75,106,124,142,178,241],"graphs)":[12],"into":[13,236],"natural":[14],"language":[15],"text.":[16],"With":[17],"the":[18,145],"exceptional":[19],"capability":[20],"of":[21,41,199],"Large":[22],"Language":[23],"Models":[24],"(LLMs),":[25],"they":[26],"have":[27],"become":[28],"ubiquitous":[29],"foundational":[31],"models":[32,134],"for":[33,43,93,138,247],"D2T.":[34,251],"This":[35,230],"article":[36],"presents":[37],"a":[38,207],"comprehensive":[39,231],"evaluation":[40,92,160,228],"LLMs":[42,64],"D2T,":[44],"focusing":[45],"on":[46,217],"three":[47,226],"key":[48,227],"qualities:":[49],"readability":[50,105],"(fluency":[51],"coherence),":[53],"informativeness":[54],"(content":[55],"preservation),":[56],"faithfulness":[58,123,143,152],"(factual":[59],"accuracy).":[60],"We":[61],"evaluate":[62],"12":[63],"from":[65,144],"five":[66,79],"prominent":[67],"open":[68],"source":[69],"families":[70],"(BART,":[71],"T5,":[72],"BLOOM,":[73],"OPT,":[74],"Llama":[76,109],"2)":[77],"across":[78,172,224],"widely":[80],"used":[81],"D2T":[82,240],"datasets":[83],"using":[84],"six":[85],"established":[86],"automatic":[87,159],"metrics,":[88],"complemented":[89],"by":[90,158],"human":[91,146],"deeper":[94],"insight.":[95],"Our":[96],"findings":[97],"reveal":[98],"that":[99,132,167,190,211],"larger":[100,133],"model":[101,117,200],"sizes":[102],"generally":[103,136],"improve":[104],"informativeness,":[107,141],"with":[108],"2":[110],"showing":[111],"superior":[112],"overall":[113],"performance.":[114],"However,":[115],"increased":[116],"size":[118],"does":[119],"not":[120],"consistently":[121,192],"enhance":[122],"may":[125],"sometimes":[126],"degrade":[127],"it.":[128],"Human":[129],"evaluations":[130],"indicate":[131],"are":[135,154],"preferred":[137],"their":[139,150,218],"readability,":[140],"readers\u2019":[147],"perspective,":[148],"minor":[151],"errors":[153],"assessed":[155],"more":[156],"selectively":[157],"metrics.":[161],"Through":[162],"robustness":[163],"analyses,":[164],"we":[165,205],"confirm":[166],"these":[168],"trends":[169],"remain":[170],"stable":[171],"different":[173],"fine-tuning":[174],"(QLoRA":[175],"vs.":[176,182],"Prefix-Tuning)":[177],"decoding":[179],"(Beam":[180],"Search":[181],"Nucleus":[183],"Sampling)":[184],"strategies.":[185],"Furthermore,":[186],"our":[187],"experiments":[188],"show":[189],"performance":[191,249],"declines":[193],"source-reference":[195,208,219],"divergence":[196],"increases,":[197],"regardless":[198],"size.":[201],"To":[202],"mitigate":[203],"this,":[204],"propose":[206],"divergence-weighted":[209],"training":[210,214,245],"adaptively":[212],"reweights":[213],"instances":[215],"based":[216],"divergence,":[220],"achieving":[221],"consistent":[222],"improvements":[223],"all":[225],"qualities.":[229],"study":[232],"provides":[233],"practical":[234],"insights":[235],"LLM":[237],"behavior":[238],"in":[239,250],"introduces":[242],"an":[243],"effective":[244],"paradigm":[246],"improving":[248]},"counts_by_year":[],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2026-02-01T00:00:00"}