{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:OONDHMZR5BK7RISMQRAIR7JDQZ","short_pith_number":"pith:OONDHMZR","schema_version":"1.0","canonical_sha256":"739a33b331e855f8a24c844088fd23867fa7e1111d99c52403ed23f556ba8556","source":{"kind":"arxiv","id":"2205.05789","version":2},"attestation_state":"computed","paper":{"title":"RITA: a Study on Scaling Up Generative Protein Sequence Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Daniel Hesslow, Debora Marks, Iacopo Poli, Niccol\\'o Zanichelli, Pascal Notin","submitted_at":"2022-05-11T22:06:03Z","abstract_excerpt":"In this work we introduce RITA: a suite of autoregressive generative models for protein sequences, with up to 1.2 billion parameters, trained on over 280 million protein sequences belonging to the UniRef-100 database. Such generative models hold the promise of greatly accelerating protein design. We conduct the first systematic study of how capabilities evolve with model size for autoregressive transformers in the protein domain: we evaluate RITA models in next amino acid prediction, zero-shot fitness, and enzyme function prediction, showing benefits from increased scale. We release the RITA m"},"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":"2205.05789","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"q-bio.QM","submitted_at":"2022-05-11T22:06:03Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8306ab7aeb6bbc10e437765a0bdc25945716b23ca8a9502852c2095512765118","abstract_canon_sha256":"9d0fd88418876beb73220e137bb796bed61d58e36432b166ec0d98b1fcaa3460"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:40:27.277623Z","signature_b64":"1GxNxDECZO3/OJJbtYY3FBtRMvLcLi4AApdhgJ7mMau3yiQQ70Rg9DSno7lcsr8kAvjgq7Mtxz0/6uf2SNNhCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"739a33b331e855f8a24c844088fd23867fa7e1111d99c52403ed23f556ba8556","last_reissued_at":"2026-07-05T04:40:27.277090Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:40:27.277090Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RITA: a Study on Scaling Up Generative Protein Sequence Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"q-bio.QM","authors_text":"Daniel Hesslow, Debora Marks, Iacopo Poli, Niccol\\'o Zanichelli, Pascal Notin","submitted_at":"2022-05-11T22:06:03Z","abstract_excerpt":"In this work we introduce RITA: a suite of autoregressive generative models for protein sequences, with up to 1.2 billion parameters, trained on over 280 million protein sequences belonging to the UniRef-100 database. Such generative models hold the promise of greatly accelerating protein design. We conduct the first systematic study of how capabilities evolve with model size for autoregressive transformers in the protein domain: we evaluate RITA models in next amino acid prediction, zero-shot fitness, and enzyme function prediction, showing benefits from increased scale. We release the RITA m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.05789","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/2205.05789/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":"2205.05789","created_at":"2026-07-05T04:40:27.277150+00:00"},{"alias_kind":"arxiv_version","alias_value":"2205.05789v2","created_at":"2026-07-05T04:40:27.277150+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.05789","created_at":"2026-07-05T04:40:27.277150+00:00"},{"alias_kind":"pith_short_12","alias_value":"OONDHMZR5BK7","created_at":"2026-07-05T04:40:27.277150+00:00"},{"alias_kind":"pith_short_16","alias_value":"OONDHMZR5BK7RISM","created_at":"2026-07-05T04:40:27.277150+00:00"},{"alias_kind":"pith_short_8","alias_value":"OONDHMZR","created_at":"2026-07-05T04:40:27.277150+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29529","citing_title":"Modeling Protein Evolution with Generative Models: from Extant Sequence Data to Evolutionary Dynamics","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27939","citing_title":"Two-Stage Fine-Tuning for Protein Sequence Generation with Targeted Amino-Acid Composition","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10189","citing_title":"ProteinOPD: Towards Effective and Efficient Preference Alignment for Protein Design","ref_index":10,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ","json":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ.json","graph_json":"https://pith.science/api/pith-number/OONDHMZR5BK7RISMQRAIR7JDQZ/graph.json","events_json":"https://pith.science/api/pith-number/OONDHMZR5BK7RISMQRAIR7JDQZ/events.json","paper":"https://pith.science/paper/OONDHMZR"},"agent_actions":{"view_html":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ","download_json":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ.json","view_paper":"https://pith.science/paper/OONDHMZR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2205.05789&json=true","fetch_graph":"https://pith.science/api/pith-number/OONDHMZR5BK7RISMQRAIR7JDQZ/graph.json","fetch_events":"https://pith.science/api/pith-number/OONDHMZR5BK7RISMQRAIR7JDQZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ/action/storage_attestation","attest_author":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ/action/author_attestation","sign_citation":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ/action/citation_signature","submit_replication":"https://pith.science/pith/OONDHMZR5BK7RISMQRAIR7JDQZ/action/replication_record"}},"created_at":"2026-07-05T04:40:27.277150+00:00","updated_at":"2026-07-05T04:40:27.277150+00:00"}