{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CJL5OODQEEOBSW3VDFLAB5ADAH","short_pith_number":"pith:CJL5OODQ","schema_version":"1.0","canonical_sha256":"1257d73870211c195b75195600f40301d22facef5ee85b5d42caf36f07cd56a9","source":{"kind":"arxiv","id":"2503.08131","version":2},"attestation_state":"computed","paper":{"title":"Large Scale Multi-Task Bayesian Optimization with Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Cesar de la Fuente-Nunez, Fangping Wan, Haydn Thomas Jones, Jacob R. Gardner, Jeffrey Tao, Marcelo Der Torossian Torres, Natalie Maus, Osbert Bastani, Ryan Marcus, Yimeng Zeng","submitted_at":"2025-03-11T07:46:19Z","abstract_excerpt":"In multi-task Bayesian optimization, the goal is to leverage experience from optimizing existing tasks to improve the efficiency of optimizing new ones. While approaches using multi-task Gaussian processes or deep kernel transfer exist, the performance improvement is marginal when scaling beyond a moderate number of tasks. We introduce a novel approach leveraging large language models (LLMs) to learn from, and improve upon, previous optimization trajectories, scaling to approximately 1500 distinct tasks. Specifically, we propose a feedback loop in which an LLM is fine-tuned on the high quality"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.08131","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-11T07:46:19Z","cross_cats_sorted":[],"title_canon_sha256":"9ec88fe9d6adc90336fadce0766ac919e344a7325423a371f837608a5ed9de53","abstract_canon_sha256":"a954ebcc4f19ee69db3debf24e473f6ff84c462ff2504de586ad991384c742f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:10.587663Z","signature_b64":"o0Fusc+Fisj6mSj5VE+YAQZ3Q2XNLW5mZXCDPpUYwNfVbtveHa2mT7OIYkEUill5Henx8DpDbC6EmS0IxQl1Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1257d73870211c195b75195600f40301d22facef5ee85b5d42caf36f07cd56a9","last_reissued_at":"2026-07-05T11:20:10.587184Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:10.587184Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Large Scale Multi-Task Bayesian Optimization with Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Cesar de la Fuente-Nunez, Fangping Wan, Haydn Thomas Jones, Jacob R. Gardner, Jeffrey Tao, Marcelo Der Torossian Torres, Natalie Maus, Osbert Bastani, Ryan Marcus, Yimeng Zeng","submitted_at":"2025-03-11T07:46:19Z","abstract_excerpt":"In multi-task Bayesian optimization, the goal is to leverage experience from optimizing existing tasks to improve the efficiency of optimizing new ones. While approaches using multi-task Gaussian processes or deep kernel transfer exist, the performance improvement is marginal when scaling beyond a moderate number of tasks. We introduce a novel approach leveraging large language models (LLMs) to learn from, and improve upon, previous optimization trajectories, scaling to approximately 1500 distinct tasks. Specifically, we propose a feedback loop in which an LLM is fine-tuned on the high quality"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.08131","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.08131/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.08131","created_at":"2026-07-05T11:20:10.587238+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.08131v2","created_at":"2026-07-05T11:20:10.587238+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.08131","created_at":"2026-07-05T11:20:10.587238+00:00"},{"alias_kind":"pith_short_12","alias_value":"CJL5OODQEEOB","created_at":"2026-07-05T11:20:10.587238+00:00"},{"alias_kind":"pith_short_16","alias_value":"CJL5OODQEEOBSW3V","created_at":"2026-07-05T11:20:10.587238+00:00"},{"alias_kind":"pith_short_8","alias_value":"CJL5OODQ","created_at":"2026-07-05T11:20:10.587238+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH","json":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH.json","graph_json":"https://pith.science/api/pith-number/CJL5OODQEEOBSW3VDFLAB5ADAH/graph.json","events_json":"https://pith.science/api/pith-number/CJL5OODQEEOBSW3VDFLAB5ADAH/events.json","paper":"https://pith.science/paper/CJL5OODQ"},"agent_actions":{"view_html":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH","download_json":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH.json","view_paper":"https://pith.science/paper/CJL5OODQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.08131&json=true","fetch_graph":"https://pith.science/api/pith-number/CJL5OODQEEOBSW3VDFLAB5ADAH/graph.json","fetch_events":"https://pith.science/api/pith-number/CJL5OODQEEOBSW3VDFLAB5ADAH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH/action/storage_attestation","attest_author":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH/action/author_attestation","sign_citation":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH/action/citation_signature","submit_replication":"https://pith.science/pith/CJL5OODQEEOBSW3VDFLAB5ADAH/action/replication_record"}},"created_at":"2026-07-05T11:20:10.587238+00:00","updated_at":"2026-07-05T11:20:10.587238+00:00"}