{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZFQOI4GSJT4FKRMV4R6IILLUUY","short_pith_number":"pith:ZFQOI4GS","schema_version":"1.0","canonical_sha256":"c960e470d24cf8554595e47c842d74a63658c2f4cafcb664575ed67d1f5eeb30","source":{"kind":"arxiv","id":"2606.26493","version":1},"attestation_state":"computed","paper":{"title":"Nemotron-TwoTower: Diffusion Language Modeling with Pretrained Autoregressive Context","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bryan Catanzaro, Fitsum Reda, John Kamalu, Mohammad Shoeybi, Mostofa Patwary, Roger Waleffe","submitted_at":"2026-06-25T00:52:44Z","abstract_excerpt":"Diffusion language models offer a promising alternative to autoregressive models due to their potential for parallel and iterative generation. However, existing approaches use a single network for both context representation and iterative denoising, forcing one model to serve both roles and limiting its capacity for either role. We propose TwoTower, a block-wise autoregressive diffusion model that decouples these roles into two towers: a frozen AR context tower that causally processes clean tokens, and a trainable diffusion denoiser tower with bidirectional block attention that refines noisy b"},"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":"2606.26493","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-25T00:52:44Z","cross_cats_sorted":[],"title_canon_sha256":"bb7c8dfb25c57555ff7ffa280aadeec1ed1e3cc2f70426f0536c1fa8ebf790bb","abstract_canon_sha256":"1ba66e1f57a46f762df602a8d503d23fc3fbe9a9c30d2d8bd694cdefbfb792a0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-26T01:15:32.711484Z","signature_b64":"Xmgto7Q4N7n+i+22DDChyh6Jrh7dsDroKYmsLPK7HxhgJV8rRgzIdSgoKVE2fFomn7EKqI2CqHPyi/2x9qRuCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c960e470d24cf8554595e47c842d74a63658c2f4cafcb664575ed67d1f5eeb30","last_reissued_at":"2026-06-26T01:15:32.711133Z","signature_status":"signed_v1","first_computed_at":"2026-06-26T01:15:32.711133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Nemotron-TwoTower: Diffusion Language Modeling with Pretrained Autoregressive Context","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bryan Catanzaro, Fitsum Reda, John Kamalu, Mohammad Shoeybi, Mostofa Patwary, Roger Waleffe","submitted_at":"2026-06-25T00:52:44Z","abstract_excerpt":"Diffusion language models offer a promising alternative to autoregressive models due to their potential for parallel and iterative generation. However, existing approaches use a single network for both context representation and iterative denoising, forcing one model to serve both roles and limiting its capacity for either role. We propose TwoTower, a block-wise autoregressive diffusion model that decouples these roles into two towers: a frozen AR context tower that causally processes clean tokens, and a trainable diffusion denoiser tower with bidirectional block attention that refines noisy b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.26493","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/2606.26493/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":"2606.26493","created_at":"2026-06-26T01:15:32.711193+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.26493v1","created_at":"2026-06-26T01:15:32.711193+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.26493","created_at":"2026-06-26T01:15:32.711193+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZFQOI4GSJT4F","created_at":"2026-06-26T01:15:32.711193+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZFQOI4GSJT4FKRMV","created_at":"2026-06-26T01:15:32.711193+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZFQOI4GS","created_at":"2026-06-26T01:15:32.711193+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/ZFQOI4GSJT4FKRMV4R6IILLUUY","json":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY.json","graph_json":"https://pith.science/api/pith-number/ZFQOI4GSJT4FKRMV4R6IILLUUY/graph.json","events_json":"https://pith.science/api/pith-number/ZFQOI4GSJT4FKRMV4R6IILLUUY/events.json","paper":"https://pith.science/paper/ZFQOI4GS"},"agent_actions":{"view_html":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY","download_json":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY.json","view_paper":"https://pith.science/paper/ZFQOI4GS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.26493&json=true","fetch_graph":"https://pith.science/api/pith-number/ZFQOI4GSJT4FKRMV4R6IILLUUY/graph.json","fetch_events":"https://pith.science/api/pith-number/ZFQOI4GSJT4FKRMV4R6IILLUUY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY/action/storage_attestation","attest_author":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY/action/author_attestation","sign_citation":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY/action/citation_signature","submit_replication":"https://pith.science/pith/ZFQOI4GSJT4FKRMV4R6IILLUUY/action/replication_record"}},"created_at":"2026-06-26T01:15:32.711193+00:00","updated_at":"2026-06-26T01:15:32.711193+00:00"}