{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:3FZYCAM252QSYPPJS3RCKLBPLT","short_pith_number":"pith:3FZYCAM2","schema_version":"1.0","canonical_sha256":"d97381019aeea12c3de996e2252c2f5ccc5572486ddf9a90f4be359a18f8b954","source":{"kind":"arxiv","id":"2306.12360","version":2},"attestation_state":"computed","paper":{"title":"Protein Discovery with Discrete Walk-Jump Sampling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.BM","authors_text":"Andreas Loukas, Daniel Berenberg, Isidro Hotzel, Joseph Kleinhenz, Julien Lafrance-Vanasse, Karina Zadorozhny, Kyunghyun Cho, Nathan C. Frey, Richard Bonneau, Saeed Saremi, Stephen Ra, Vladimir Gligorijevic, Yan Wu","submitted_at":"2023-06-08T17:03:46Z","abstract_excerpt":"We resolve difficulties in training and sampling from a discrete generative model by learning a smoothed energy function, sampling from the smoothed data manifold with Langevin Markov chain Monte Carlo (MCMC), and projecting back to the true data manifold with one-step denoising. Our Discrete Walk-Jump Sampling formalism combines the contrastive divergence training of an energy-based model and improved sample quality of a score-based model, while simplifying training and sampling by requiring only a single noise level. We evaluate the robustness of our approach on generative modeling of antibo"},"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":"2306.12360","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"q-bio.BM","submitted_at":"2023-06-08T17:03:46Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0c65649427d252bb6ea7dddea9781eb88aa057f57381d81613f761d0cdd1ff0f","abstract_canon_sha256":"f13aedbbd27c876a512263f5e10df274af25422e47a73a8b16279be8a43460ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:30:51.042852Z","signature_b64":"zupj/QhamVOSs58jX+PrpADojKwrp0SHM1/5IrOJgamDx3Vxpm/DW9efWwJISOQ8kJuN9k09XshEkAb8KC28CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d97381019aeea12c3de996e2252c2f5ccc5572486ddf9a90f4be359a18f8b954","last_reissued_at":"2026-07-05T10:30:51.042287Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:30:51.042287Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Protein Discovery with Discrete Walk-Jump Sampling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.BM","authors_text":"Andreas Loukas, Daniel Berenberg, Isidro Hotzel, Joseph Kleinhenz, Julien Lafrance-Vanasse, Karina Zadorozhny, Kyunghyun Cho, Nathan C. Frey, Richard Bonneau, Saeed Saremi, Stephen Ra, Vladimir Gligorijevic, Yan Wu","submitted_at":"2023-06-08T17:03:46Z","abstract_excerpt":"We resolve difficulties in training and sampling from a discrete generative model by learning a smoothed energy function, sampling from the smoothed data manifold with Langevin Markov chain Monte Carlo (MCMC), and projecting back to the true data manifold with one-step denoising. Our Discrete Walk-Jump Sampling formalism combines the contrastive divergence training of an energy-based model and improved sample quality of a score-based model, while simplifying training and sampling by requiring only a single noise level. We evaluate the robustness of our approach on generative modeling of antibo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.12360","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/2306.12360/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":"2306.12360","created_at":"2026-07-05T10:30:51.042360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.12360v2","created_at":"2026-07-05T10:30:51.042360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.12360","created_at":"2026-07-05T10:30:51.042360+00:00"},{"alias_kind":"pith_short_12","alias_value":"3FZYCAM252QS","created_at":"2026-07-05T10:30:51.042360+00:00"},{"alias_kind":"pith_short_16","alias_value":"3FZYCAM252QSYPPJ","created_at":"2026-07-05T10:30:51.042360+00:00"},{"alias_kind":"pith_short_8","alias_value":"3FZYCAM2","created_at":"2026-07-05T10:30:51.042360+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/3FZYCAM252QSYPPJS3RCKLBPLT","json":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT.json","graph_json":"https://pith.science/api/pith-number/3FZYCAM252QSYPPJS3RCKLBPLT/graph.json","events_json":"https://pith.science/api/pith-number/3FZYCAM252QSYPPJS3RCKLBPLT/events.json","paper":"https://pith.science/paper/3FZYCAM2"},"agent_actions":{"view_html":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT","download_json":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT.json","view_paper":"https://pith.science/paper/3FZYCAM2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.12360&json=true","fetch_graph":"https://pith.science/api/pith-number/3FZYCAM252QSYPPJS3RCKLBPLT/graph.json","fetch_events":"https://pith.science/api/pith-number/3FZYCAM252QSYPPJS3RCKLBPLT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT/action/storage_attestation","attest_author":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT/action/author_attestation","sign_citation":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT/action/citation_signature","submit_replication":"https://pith.science/pith/3FZYCAM252QSYPPJS3RCKLBPLT/action/replication_record"}},"created_at":"2026-07-05T10:30:51.042360+00:00","updated_at":"2026-07-05T10:30:51.042360+00:00"}