{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VKNXPM42IQV4TLUBKAVN2T2SAF","short_pith_number":"pith:VKNXPM42","schema_version":"1.0","canonical_sha256":"aa9b77b39a442bc9ae81502add4f520175ffa7a68ab7328f5cdcf733768d75d4","source":{"kind":"arxiv","id":"2507.03910","version":1},"attestation_state":"computed","paper":{"title":"Return of the Latent Space COWBOYS: Re-thinking the use of VAEs for Bayesian Optimisation of Structured Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Henry B. Moss, Sebastian W. Ober, Tom Diethe","submitted_at":"2025-07-05T05:53:04Z","abstract_excerpt":"Bayesian optimisation in the latent space of a Variational AutoEncoder (VAE) is a powerful framework for optimisation tasks over complex structured domains, such as the space of scientifically interesting molecules. However, existing approaches tightly couple the surrogate and generative models, which can lead to suboptimal performance when the latent space is not tailored to specific tasks, which in turn has led to the proposal of increasingly sophisticated algorithms. In this work, we explore a new direction, instead proposing a decoupled approach that trains a generative model and a Gaussia"},"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":"2507.03910","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-05T05:53:04Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"62f5cbc8839d8ff1cb934d40a45bac24843311be748e4be9fe6aad89b5ea595a","abstract_canon_sha256":"41fb6f9b6edfc56e3a66b2be93a1cd0300b375b69af85ec2c2611d79fd5a969d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:32:37.138803Z","signature_b64":"6AtCn3XpG3woDsouEOL8+lFvccAjgj0mXeIcwUJx5YGCZ7Svc/Pc2OMvwzsQmDre1BLivWoGarI8SwpPbZNWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa9b77b39a442bc9ae81502add4f520175ffa7a68ab7328f5cdcf733768d75d4","last_reissued_at":"2026-07-05T11:32:37.138285Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:32:37.138285Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Return of the Latent Space COWBOYS: Re-thinking the use of VAEs for Bayesian Optimisation of Structured Spaces","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Henry B. Moss, Sebastian W. Ober, Tom Diethe","submitted_at":"2025-07-05T05:53:04Z","abstract_excerpt":"Bayesian optimisation in the latent space of a Variational AutoEncoder (VAE) is a powerful framework for optimisation tasks over complex structured domains, such as the space of scientifically interesting molecules. However, existing approaches tightly couple the surrogate and generative models, which can lead to suboptimal performance when the latent space is not tailored to specific tasks, which in turn has led to the proposal of increasingly sophisticated algorithms. In this work, we explore a new direction, instead proposing a decoupled approach that trains a generative model and a Gaussia"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03910","kind":"arxiv","version":1},"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/2507.03910/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":"2507.03910","created_at":"2026-07-05T11:32:37.138365+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.03910v1","created_at":"2026-07-05T11:32:37.138365+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03910","created_at":"2026-07-05T11:32:37.138365+00:00"},{"alias_kind":"pith_short_12","alias_value":"VKNXPM42IQV4","created_at":"2026-07-05T11:32:37.138365+00:00"},{"alias_kind":"pith_short_16","alias_value":"VKNXPM42IQV4TLUB","created_at":"2026-07-05T11:32:37.138365+00:00"},{"alias_kind":"pith_short_8","alias_value":"VKNXPM42","created_at":"2026-07-05T11:32:37.138365+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.19093","citing_title":"Embedding by Elicitation: Dynamic Representations for Bayesian Optimization of System Prompts","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF","json":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF.json","graph_json":"https://pith.science/api/pith-number/VKNXPM42IQV4TLUBKAVN2T2SAF/graph.json","events_json":"https://pith.science/api/pith-number/VKNXPM42IQV4TLUBKAVN2T2SAF/events.json","paper":"https://pith.science/paper/VKNXPM42"},"agent_actions":{"view_html":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF","download_json":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF.json","view_paper":"https://pith.science/paper/VKNXPM42","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.03910&json=true","fetch_graph":"https://pith.science/api/pith-number/VKNXPM42IQV4TLUBKAVN2T2SAF/graph.json","fetch_events":"https://pith.science/api/pith-number/VKNXPM42IQV4TLUBKAVN2T2SAF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF/action/storage_attestation","attest_author":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF/action/author_attestation","sign_citation":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF/action/citation_signature","submit_replication":"https://pith.science/pith/VKNXPM42IQV4TLUBKAVN2T2SAF/action/replication_record"}},"created_at":"2026-07-05T11:32:37.138365+00:00","updated_at":"2026-07-05T11:32:37.138365+00:00"}