{"id":"https://openalex.org/W7018961719","doi":null,"title":"Evaluating the Performance of Direct Preference Optimization in News Article Title Generation : A Comparative Analysis with Supervised Fine-Tuning","display_name":"Evaluating the Performance of Direct Preference Optimization in News Article Title Generation : A Comparative Analysis with Supervised Fine-Tuning","publication_year":2024,"publication_date":"2024-01-01","ids":{"openalex":"https://openalex.org/W7018961719"},"language":"en","primary_location":{"id":"pmh:oai:DiVA.org:kth-352816","doi":null,"is_oa":true,"landing_page_url":"http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-352816","pdf_url":"https://kth.diva-portal.org/smash/get/diva2:1895800/FULLTEXT01","source":{"id":"https://openalex.org/S4306401559","display_name":"KTH Publication Database DiVA (KTH Royal Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I86987016","host_organization_name":"KTH Royal Institute of Technology","host_organization_lineage":["https://openalex.org/I86987016"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/bachelorThesis"},"volume":null,"type":"dissertation","is_traditional":true,"indexed_in":[],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://kth.diva-portal.org/smash/get/diva2:1895800/FULLTEXT01","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Dettner K\u00e4llander, Hugo","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Dettner K\u00e4llander, Hugo","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":true,"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":true,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.9282000064849854,"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":0.9282000064849854,"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":0.034299999475479126,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.006099999882280827,"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/direct-preference-optimization","display_name":"direct preference optimization","score":0.8429999947547913},{"id":"https://openalex.org/keywords/supervised-fine-tuning","display_name":"supervised fine-tuning","score":0.9980000257492065},{"id":"https://openalex.org/keywords/large-language-models","display_name":"large language models","score":0.9620000123977661},{"id":"https://openalex.org/keywords/human-preference-data","display_name":"human preference data","score":0.5379999876022339},{"id":"https://openalex.org/keywords/model-selection","display_name":"model selection","score":0.5899999737739563},{"id":"https://openalex.org/keywords/a-b-testing","display_name":"A/B testing","score":0.6230000257492065}],"concepts":[{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5773000121116638},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5741999745368958},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5679000020027161},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5641999840736389},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5309000015258789},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.5300999879837036},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.5295000076293945},{"id":"https://openalex.org/C2778334786","wikidata":"https://www.wikidata.org/wiki/Q1586270","display_name":"Variation (astronomy)","level":2,"score":0.5077999830245972},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.48019999265670776},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.46889999508857727},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.44600000977516174},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.3190999925136566},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.30709999799728394},{"id":"https://openalex.org/C189430467","wikidata":"https://www.wikidata.org/wiki/Q7293293","display_name":"Ranking (information retrieval)","level":2,"score":0.27649998664855957},{"id":"https://openalex.org/C181204326","wikidata":"https://www.wikidata.org/wiki/Q7239820","display_name":"Preference learning","level":3,"score":0.272599995136261},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"pmh:oai:DiVA.org:kth-352816","doi":null,"is_oa":true,"landing_page_url":"http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-352816","pdf_url":"https://kth.diva-portal.org/smash/get/diva2:1895800/FULLTEXT01","source":{"id":"https://openalex.org/S4306401559","display_name":"KTH Publication Database DiVA (KTH Royal Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I86987016","host_organization_name":"KTH Royal Institute of Technology","host_organization_lineage":["https://openalex.org/I86987016"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/bachelorThesis"}],"best_oa_location":{"id":"pmh:oai:DiVA.org:kth-352816","doi":null,"is_oa":true,"landing_page_url":"http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-352816","pdf_url":"https://kth.diva-portal.org/smash/get/diva2:1895800/FULLTEXT01","source":{"id":"https://openalex.org/S4306401559","display_name":"KTH Publication Database DiVA (KTH Royal Institute of Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I86987016","host_organization_name":"KTH Royal Institute of Technology","host_organization_lineage":["https://openalex.org/I86987016"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-sa","license_id":"https://openalex.org/licenses/cc-by-sa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"info:eu-repo/semantics/bachelorThesis"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7018961719.pdf","grobid_xml":"https://content.openalex.org/works/W7018961719.grobid-xml"},"referenced_works_count":31,"referenced_works":["https://openalex.org/W1565746575","https://openalex.org/W1570639452","https://openalex.org/W2141403362","https://openalex.org/W2327037637","https://openalex.org/W2507324087","https://openalex.org/W2626778328","https://openalex.org/W2626804490","https://openalex.org/W2885088125","https://openalex.org/W2919115771","https://openalex.org/W2973379954","https://openalex.org/W3030163527","https://openalex.org/W3035725536","https://openalex.org/W3082115681","https://openalex.org/W3118781290","https://openalex.org/W3125925269","https://openalex.org/W3210130067","https://openalex.org/W4212863985","https://openalex.org/W4362597869","https://openalex.org/W4377121527","https://openalex.org/W4386576685","https://openalex.org/W4386793209","https://openalex.org/W4387801352","https://openalex.org/W4388979610","https://openalex.org/W4389519192","https://openalex.org/W4389520260","https://openalex.org/W4390573302","https://openalex.org/W4391494845","https://openalex.org/W4391555989","https://openalex.org/W4400414492","https://openalex.org/W4401042394","https://openalex.org/W7133208539"],"related_works":["https://openalex.org/W7164498433","https://openalex.org/W4403704388","https://openalex.org/W4392019816","https://openalex.org/W6891904700","https://openalex.org/W7164548507","https://openalex.org/W7162107464","https://openalex.org/W4400667828","https://openalex.org/W7140123314","https://openalex.org/W4394973490","https://openalex.org/W4412888233"],"abstract_inverted_index":{"Typically,":[0],"Large":[1],"Language":[2],"Models":[3],"(LLMs)":[4],"are":[5,97],"adapted":[6],"to":[7,23,55,61,87,112,184,202,244],"a":[8,36,77,125,204],"specific":[9],"task":[10],"using":[11,114,124,259],"large":[12,102],"amounts":[13],"of":[14,65,73,91,104,129,166,188],"labeled":[15],"training":[16,38],"data.":[17],"This":[18,58,118,177],"process":[19],"has":[20,51],"been":[21,32,53,209],"found":[22,210],"teach":[24],"the":[25,63,88,92,101,147,163,167,186,197,248],"LLM":[26,173],"undesirable":[27],"behavior,":[28],"an":[29,236],"issue":[30],"that\u2019s":[31],"effectively":[33],"solved":[34],"by":[35,123,154,231,235],"recent":[37],"method":[39,50,198],"incorporating":[40],"human":[41,116,175,232],"preference,":[42],"called":[43],"Direct":[44],"Preference":[45],"Optimization":[46],"(DPO).":[47],"However,":[48,85],"this":[49,141,143],"largely":[52],"limited":[54],"instruction-based":[56],"LLMs.":[57],"thesis":[59,119,144],"seeks":[60],"evaluate":[62,113],"performance":[64],"Supervised":[66],"Fine-Tuning":[67],"(SFT)":[68],"and":[69,71,100,107,132,138,227,250],"DPO":[70,105,150,181,226],"several":[72],"its":[74],"variations":[75,106],"in":[76,171,211,219,247],"new,":[78],"non-instruction-based":[79],"context:":[80],"news":[81],"article":[82,190],"title":[83],"generation.":[84],"due":[86],"subjective":[89,133],"nature":[90],"task,":[93],"simple":[94],"objective":[95,131,155],"measures":[96,156],"not":[98,158],"enough,":[99],"number":[103],"hyperparameter":[108],"values":[109],"is":[110,229],"infeasible":[111],"only":[115],"evaluators.":[117],"solves":[120],"these":[121,239,252],"issues":[122],"multi-step":[126],"pipeline":[127],"consisting":[128],"both":[130],"metrics":[134],"for":[135],"model":[136,170],"selection":[137],"evaluation.":[139],"Using":[140],"method,":[142],"finds":[145],"that":[146,179,196,206,218],"most":[148],"performant":[149],"variation":[151],"as":[152],"determined":[153],"did":[157],"perform":[159],"significantly":[160],"differently":[161],"from":[162],"SFT":[164,228],"variant":[165],"same":[168],"base":[169],"either":[172,180],"or":[174,195,257],"evaluations.":[176],"indicates":[178],"was":[182,200],"unable":[183,201],"learn":[185],"patterns":[187],"successful":[189],"titles":[191],"better":[192],"than":[193],"SFT,":[194],"used":[199],"detect":[203],"difference":[205,224],"would":[207],"have":[208],"real-world":[212],"A/B":[213,261],"testing.":[214,262],"These":[215],"results":[216],"show":[217],"some":[220],"tasks":[221],"no":[222],"quality":[223],"between":[225],"detectable":[230],"experts":[233],"nor":[234],"LLM.":[237],"Hopefully,":[238],"conclusions":[240,253],"will":[241],"be":[242,255],"helpful":[243],"future":[245],"research":[246],"area,":[249],"perhaps":[251],"could":[254],"confirmed":[256],"rejected":[258],"proper":[260]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-10T00:00:00"}