{"id":"https://openalex.org/W7161845052","doi":"https://doi.org/10.48550/arxiv.2605.19093","title":"Embedding by Elicitation: Dynamic Representations for Bayesian Optimization of System Prompts","display_name":"Embedding by Elicitation: Dynamic Representations for Bayesian Optimization of System Prompts","publication_year":2026,"publication_date":"2026-05-18","ids":{"openalex":"https://openalex.org/W7161845052","doi":"https://doi.org/10.48550/arxiv.2605.19093"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.19093","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19093","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.19093","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5057291277","display_name":"Zhiyuan Lin","orcid":"https://orcid.org/0000-0003-3739-769X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Zhiyuan Jerry","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043715747","display_name":"Benjamin Letham","orcid":"https://orcid.org/0009-0003-5205-8212"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Letham, Benjamin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124612786","display_name":"Samuel Dooley","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dooley, Samuel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055451857","display_name":"Maximilian Balandat","orcid":"https://orcid.org/0000-0002-8214-8935"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Balandat, Maximilian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5006700143","display_name":"Eytan Bakshy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bakshy, Eytan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":5,"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/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.17239999771118164,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},"topics":[{"id":"https://openalex.org/T12101","display_name":"Advanced Bandit Algorithms Research","score":0.17239999771118164,"subfield":{"id":"https://openalex.org/subfields/1803","display_name":"Management Science and Operations Research"},"field":{"id":"https://openalex.org/fields/18","display_name":"Decision Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}},{"id":"https://openalex.org/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.1331000030040741,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.07119999825954437,"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/bayesian-optimization","display_name":"Bayesian optimization","score":0.7641000151634216},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5799999833106995},{"id":"https://openalex.org/keywords/aggregate","display_name":"Aggregate (composite)","score":0.5752999782562256},{"id":"https://openalex.org/keywords/probabilistic-logic","display_name":"Probabilistic logic","score":0.5670999884605408},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5619999766349792},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.550000011920929},{"id":"https://openalex.org/keywords/gaussian-process","display_name":"Gaussian process","score":0.5138999819755554},{"id":"https://openalex.org/keywords/bayesian-probability","display_name":"Bayesian probability","score":0.501800000667572},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.5005999803543091}],"concepts":[{"id":"https://openalex.org/C2778049539","wikidata":"https://www.wikidata.org/wiki/Q17002908","display_name":"Bayesian optimization","level":2,"score":0.7641000151634216},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6974999904632568},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6189000010490417},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5799999833106995},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.5752999782562256},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5728999972343445},{"id":"https://openalex.org/C49937458","wikidata":"https://www.wikidata.org/wiki/Q2599292","display_name":"Probabilistic logic","level":2,"score":0.5670999884605408},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5619999766349792},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.550000011920929},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.5138999819755554},{"id":"https://openalex.org/C107673813","wikidata":"https://www.wikidata.org/wiki/Q812534","display_name":"Bayesian probability","level":2,"score":0.501800000667572},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5005999803543091},{"id":"https://openalex.org/C83665646","wikidata":"https://www.wikidata.org/wiki/Q42139305","display_name":"Feature vector","level":2,"score":0.45500001311302185},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.43320000171661377},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3846000134944916},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.358599990606308},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3353999853134155},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.3269999921321869},{"id":"https://openalex.org/C160234255","wikidata":"https://www.wikidata.org/wiki/Q812535","display_name":"Bayesian inference","level":3,"score":0.3167000114917755},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.31209999322891235},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3093000054359436},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.3075000047683716},{"id":"https://openalex.org/C207201462","wikidata":"https://www.wikidata.org/wiki/Q182505","display_name":"Bayes' theorem","level":3,"score":0.3075000047683716},{"id":"https://openalex.org/C57691317","wikidata":"https://www.wikidata.org/wiki/Q1289248","display_name":"Scalar (mathematics)","level":2,"score":0.3057999908924103},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.29499998688697815},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2939999997615814},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.28529998660087585},{"id":"https://openalex.org/C2778827112","wikidata":"https://www.wikidata.org/wiki/Q22245680","display_name":"Feature engineering","level":3,"score":0.2705000042915344},{"id":"https://openalex.org/C55660270","wikidata":"https://www.wikidata.org/wiki/Q5164377","display_name":"Constrained optimization","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C183115368","wikidata":"https://www.wikidata.org/wiki/Q856577","display_name":"Weighting","level":2,"score":0.2653000056743622}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.19093","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19093","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.19093","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.19093","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":"article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"System":[0],"prompts":[1,86],"are":[2,21],"a":[3,57,67,79,90,139,165],"central":[4],"control":[5],"mechanism":[6],"in":[7],"modern":[8],"AI":[9],"systems,":[10],"shaping":[11],"behavior":[12],"across":[13],"conversations,":[14],"tasks,":[15],"and":[16,73,84,107,151],"user":[17],"populations.":[18],"Yet":[19],"they":[20],"difficult":[22],"to":[23,125],"tune":[24],"when":[25],"feedback":[26,44],"is":[27],"available":[28],"only":[29,195],"as":[30,46,117,138,189],"aggregate":[31,43,141,174],"metrics":[32],"rather":[33],"than":[34],"per-example":[35,153],"labels,":[36,154],"failures,":[37],"or":[38,156],"critiques.":[39,157],"We":[40,54,130],"study":[41],"this":[42],"setting":[45,133],"sample-constrained":[47],"black-box":[48],"optimization":[49,59,162,200],"over":[50,201],"discrete,":[51],"variable-length":[52],"text.":[53],"introduce":[55],"ReElicit,":[56],"Bayesian":[58,199],"framework":[60],"based":[61],"on":[62],"\\emph{embedding":[63],"by":[64],"elicitation}.":[65],"Given":[66],"task":[68],"description,":[69],"previously":[70],"evaluated":[71],"prompts,":[72],"scalar":[74,147],"scores,":[75],"an":[76,95],"LLM":[77,105],"elicits":[78],"compact,":[80],"interpretable":[81],"feature":[82,101,115],"space":[83,116],"maps":[85],"into":[87,109],"it.":[88],"Leveraging":[89],"probabilistic":[91],"Gaussian":[92],"process":[93],"surrogate,":[94],"acquisition":[96],"function":[97],"then":[98],"selects":[99],"target":[100],"vectors,":[102],"which":[103],"the":[104,114,122,126,132,143,172],"realizes":[106],"refines":[108],"deployable":[110],"system":[111,160],"prompts.":[112],"Re-eliciting":[113],"new":[118],"evaluations":[119],"arrive":[120],"lets":[121],"representation":[123,192],"adapt":[124],"observed":[127],"prompt-score":[128],"history.":[129],"evaluate":[131],"using":[134],"offline":[135],"benchmark":[136],"accuracy":[137],"controlled":[140],"proxy:":[142],"optimizer":[144],"observes":[145],"one":[146],"score":[148],"per":[149],"prompt":[150,161,196],"no":[152],"errors,":[155],"Across":[158],"ten":[159],"tasks":[163],"with":[164],"30":[166],"total":[167],"evaluation":[168],"budget,":[169],"ReElicit":[170],"achieves":[171],"strongest":[173],"performance":[175],"profile":[176],"among":[177],"representative":[178],"aggregate-only":[179],"prompt-optimization":[180],"baselines.":[181],"These":[182],"results":[183],"suggest":[184],"that":[185],"LLMs":[186],"can":[187],"serve":[188],"adaptive":[190],"semantic":[191],"builders,":[193],"not":[194],"generators,":[197],"for":[198],"natural-language":[202],"artifacts.":[203]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-21T00:00:00"}