{"id":"https://openalex.org/W4412888733","doi":"https://doi.org/10.18653/v1/2025.findings-acl.147","title":"Model Performance-Guided Evaluation Data Selection for Effective Prompt Optimization","display_name":"Model Performance-Guided Evaluation Data Selection for Effective Prompt Optimization","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4412888733","doi":"https://doi.org/10.18653/v1/2025.findings-acl.147"},"language":"en","primary_location":{"id":"doi:10.18653/v1/2025.findings-acl.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.147","pdf_url":"https://aclanthology.org/2025.findings-acl.147.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://aclanthology.org/2025.findings-acl.147.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5113198167","display_name":"Ximing Dong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ximing Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100664833","display_name":"Shaowei Wang","orcid":"https://orcid.org/0000-0003-3823-1771"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shaowei Wang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101348461","display_name":"Dayi Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dayi Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5112961078","display_name":"Ahmed Hassan","orcid":"https://orcid.org/0009-0004-2584-563X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ahmed Hassan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":{"value":0.18900461,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2844","last_page":"2859"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12127","display_name":"Software System Performance and Reliability","score":0.928600013256073,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/T12127","display_name":"Software System Performance and Reliability","score":0.928600013256073,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/computer-science","display_name":"Computer science","score":0.6987481117248535},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.6733584403991699},{"id":"https://openalex.org/keywords/data-modeling","display_name":"Data modeling","score":0.4175987243652344},{"id":"https://openalex.org/keywords/reliability-engineering","display_name":"Reliability engineering","score":0.34753018617630005},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.22777920961380005},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.12924811244010925},{"id":"https://openalex.org/keywords/database","display_name":"Database","score":0.10374525189399719}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6987481117248535},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.6733584403991699},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.4175987243652344},{"id":"https://openalex.org/C200601418","wikidata":"https://www.wikidata.org/wiki/Q2193887","display_name":"Reliability engineering","level":1,"score":0.34753018617630005},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.22777920961380005},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.12924811244010925},{"id":"https://openalex.org/C77088390","wikidata":"https://www.wikidata.org/wiki/Q8513","display_name":"Database","level":1,"score":0.10374525189399719}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.18653/v1/2025.findings-acl.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.147","pdf_url":"https://aclanthology.org/2025.findings-acl.147.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.18653/v1/2025.findings-acl.147","is_oa":true,"landing_page_url":"https://doi.org/10.18653/v1/2025.findings-acl.147","pdf_url":"https://aclanthology.org/2025.findings-acl.147.pdf","source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Findings of the Association for Computational Linguistics: ACL 2025","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4412888733.pdf","grobid_xml":"https://content.openalex.org/works/W4412888733.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"Optimizing":[0],"Large":[1],"Language":[2],"Model":[3,100],"(LLM)":[4],"performance":[5,76,125],"requires":[6],"well-crafted":[7],"prompts,":[8],"but":[9,24],"manual":[10],"prompt":[11,18,59],"engineering":[12],"is":[13,102],"labor-intensive":[14],"and":[15,46,72,109,115,135,137,141,154,171],"often":[16],"ineffective.Automated":[17],"optimization":[19,60],"techniques":[20],"address":[21],"this":[22],"challenge":[23],"the":[25,39,73,164,168,179],"majority":[26],"of":[27,75],"them":[28],"rely":[29],"on":[30,131],"randomly":[31],"selected":[32],"evaluation":[33,91],"subsets,":[34],"which":[35],"fail":[36],"to":[37,43,62,127,152,160,192],"represent":[38],"full":[40],"dataset,":[41],"leading":[42],"unreliable":[44],"evaluations":[45],"suboptimal":[47],"prompts.Existing":[48],"coreset":[49,195],"selection":[50,93,196],"methods,":[51],"designed":[52],"for":[53,58,78,94],"LLM":[54],"benchmarking,":[55],"are":[56],"unsuitable":[57],"due":[61],"challenges":[63],"in":[64],"clustering":[65,114],"similar":[66],"samples,":[67],"high":[68],"data":[69,77,92,126],"collection":[70],"costs,":[71],"unavailability":[74],"new":[79],"or":[80],"private":[81],"datasets.To":[82],"overcome":[83],"these":[84],"issues,":[85],"we":[86],"propose":[87],"IPOMP,":[88],"an":[89],"Iterative":[90],"effective":[95],"Prompt":[96],"Optimization":[97],"using":[98,112],"real-time":[99,123,184],"Performance.IPOMP":[101],"a":[103],"two-stage":[104],"approach":[105,187],"that":[106,144,182],"selects":[107],"representative":[108],"diverse":[110],"samples":[111],"semantic":[113],"boundary":[116],"analysis,":[117],"followed":[118],"by":[119,148,156],"iterative":[120],"refinement":[121,186],"with":[122,163,173],"model":[124],"replace":[128],"redundant":[129],"samples.Evaluations":[130],"two":[132,138],"datasets":[133,170],"BIG-bench":[134],"LIAR,":[136],"models":[139],"GPT-3.5":[140],"GPT-4omini,":[142],"show":[143],"IPOMP":[145],"improves":[146],"effectiveness":[147],"at":[149,157],"least":[150,158],"1.6%":[151],"3.1%,":[153],"stability":[155],"50%":[159],"55.5%":[161],"compared":[162],"best":[165],"baseline":[166],"across":[167],"studied":[169],"models,":[172],"minimal":[174],"computational":[175],"overhead":[176],"below":[177],"1%.Furthermore,":[178],"results":[180],"demonstrate":[181],"our":[183],"performance-guided":[185],"can":[188],"be":[189],"universally":[190],"applied":[191],"enhance":[193],"existing":[194],"methods.":[197]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}