{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:DZHXP5TPY5I4OJPDBQHDICWAPT","short_pith_number":"pith:DZHXP5TP","schema_version":"1.0","canonical_sha256":"1e4f77f66fc751c725e30c0e340ac07cd7b3b1df17b8b81a55b47e22cacc61e7","source":{"kind":"arxiv","id":"2106.04399","version":2},"attestation_state":"computed","paper":{"title":"Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Doina Precup, Emmanuel Bengio, Maksym Korablyov, Moksh Jain, Yoshua Bengio","submitted_at":"2021-06-08T14:21:10Z","abstract_excerpt":"This paper is about the problem of learning a stochastic policy for generating an object (like a molecular graph) from a sequence of actions, such that the probability of generating an object is proportional to a given positive reward for that object. Whereas standard return maximization tends to converge to a single return-maximizing sequence, there are cases where we would like to sample a diverse set of high-return solutions. These arise, for example, in black-box function optimization when few rounds are possible, each with large batches of queries, where the batches should be diverse, e.g"},"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":"2106.04399","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-06-08T14:21:10Z","cross_cats_sorted":[],"title_canon_sha256":"1904269d4b629716150d8f3543856d56eab047a061137a1f151814336935df26","abstract_canon_sha256":"0678de37659e6e63c5bd2c696234c600423ec29e5068498c46bce8994bcd5525"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:33:22.058794Z","signature_b64":"kUgq9znC7Upb2aQppZ3PkLfoPnL3bqvexZ/AZB7QJ+3sG4ABOoCcvZcZsC4yufGeKaT8CwI4kXJtAweL49dmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e4f77f66fc751c725e30c0e340ac07cd7b3b1df17b8b81a55b47e22cacc61e7","last_reissued_at":"2026-07-05T03:33:22.058287Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:33:22.058287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Flow Network based Generative Models for Non-Iterative Diverse Candidate Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Doina Precup, Emmanuel Bengio, Maksym Korablyov, Moksh Jain, Yoshua Bengio","submitted_at":"2021-06-08T14:21:10Z","abstract_excerpt":"This paper is about the problem of learning a stochastic policy for generating an object (like a molecular graph) from a sequence of actions, such that the probability of generating an object is proportional to a given positive reward for that object. Whereas standard return maximization tends to converge to a single return-maximizing sequence, there are cases where we would like to sample a diverse set of high-return solutions. These arise, for example, in black-box function optimization when few rounds are possible, each with large batches of queries, where the batches should be diverse, e.g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.04399","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/2106.04399/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":"2106.04399","created_at":"2026-07-05T03:33:22.058356+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.04399v2","created_at":"2026-07-05T03:33:22.058356+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.04399","created_at":"2026-07-05T03:33:22.058356+00:00"},{"alias_kind":"pith_short_12","alias_value":"DZHXP5TPY5I4","created_at":"2026-07-05T03:33:22.058356+00:00"},{"alias_kind":"pith_short_16","alias_value":"DZHXP5TPY5I4OJPD","created_at":"2026-07-05T03:33:22.058356+00:00"},{"alias_kind":"pith_short_8","alias_value":"DZHXP5TP","created_at":"2026-07-05T03:33:22.058356+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.26540","citing_title":"DGLD: Domain-Gated Latent Diffusion for the Discovery of Novel Energetic Materials","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16797","citing_title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.02300","citing_title":"A Meta Reinforcement Learning Approach to Goals-Based Wealth Management","ref_index":258,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT","json":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT.json","graph_json":"https://pith.science/api/pith-number/DZHXP5TPY5I4OJPDBQHDICWAPT/graph.json","events_json":"https://pith.science/api/pith-number/DZHXP5TPY5I4OJPDBQHDICWAPT/events.json","paper":"https://pith.science/paper/DZHXP5TP"},"agent_actions":{"view_html":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT","download_json":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT.json","view_paper":"https://pith.science/paper/DZHXP5TP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.04399&json=true","fetch_graph":"https://pith.science/api/pith-number/DZHXP5TPY5I4OJPDBQHDICWAPT/graph.json","fetch_events":"https://pith.science/api/pith-number/DZHXP5TPY5I4OJPDBQHDICWAPT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT/action/storage_attestation","attest_author":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT/action/author_attestation","sign_citation":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT/action/citation_signature","submit_replication":"https://pith.science/pith/DZHXP5TPY5I4OJPDBQHDICWAPT/action/replication_record"}},"created_at":"2026-07-05T03:33:22.058356+00:00","updated_at":"2026-07-05T03:33:22.058356+00:00"}