{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LUNAXDW4RKFA3QRNDJPCEHLLNA","short_pith_number":"pith:LUNAXDW4","schema_version":"1.0","canonical_sha256":"5d1a0b8edc8a8a0dc22d1a5e221d6b681dabad1cbef83d41efaab8bc885e0486","source":{"kind":"arxiv","id":"2402.19097","version":4},"attestation_state":"computed","paper":{"title":"TEncDM: Understanding the Properties of the Diffusion Model in the Space of Language Model Encodings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander Shabalin, Dmitry Molchanov, Dmitry Vetrov, Egor Chimbulatov, Grigory Bartosh, Roman Kim, Sergey Markov, Viacheslav Meshchaninov, Vladislav Lapikov","submitted_at":"2024-02-29T12:25:45Z","abstract_excerpt":"This paper presents the Text Encoding Diffusion Model (TEncDM), a novel approach to diffusion modeling that operates in the space of pre-trained language model encodings. In contrast to traditionally used embeddings, encodings integrate contextual information. In our approach, we also employ a transformer-based decoder, specifically designed to incorporate context in the token prediction process. We conduct a comprehensive examination of the influence of the encoder, decoder, noise scheduler, and self-conditioning on zero-shot generation. Furthermore, we compare TEncDM with previous approaches"},"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":"2402.19097","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-29T12:25:45Z","cross_cats_sorted":[],"title_canon_sha256":"aff6bd96edc1904f9594b27fad04a472dec7da6b07f84ece7c3bc303fd9e7d32","abstract_canon_sha256":"95815af0333e2b9922073ddd5d37dabcdb125fa87ff10c15eb8c4d195af8b00c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:18:35.179225Z","signature_b64":"6bRVOHSydoQ4Zr41aQcvwuB0PH8shwY7BtgKhe20xt7DkxbIPkpYh9MjNRu77OPBAeTwLI4tGvYhBeGOM7ZiAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d1a0b8edc8a8a0dc22d1a5e221d6b681dabad1cbef83d41efaab8bc885e0486","last_reissued_at":"2026-07-05T10:18:35.178737Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:18:35.178737Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TEncDM: Understanding the Properties of the Diffusion Model in the Space of Language Model Encodings","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexander Shabalin, Dmitry Molchanov, Dmitry Vetrov, Egor Chimbulatov, Grigory Bartosh, Roman Kim, Sergey Markov, Viacheslav Meshchaninov, Vladislav Lapikov","submitted_at":"2024-02-29T12:25:45Z","abstract_excerpt":"This paper presents the Text Encoding Diffusion Model (TEncDM), a novel approach to diffusion modeling that operates in the space of pre-trained language model encodings. In contrast to traditionally used embeddings, encodings integrate contextual information. In our approach, we also employ a transformer-based decoder, specifically designed to incorporate context in the token prediction process. We conduct a comprehensive examination of the influence of the encoder, decoder, noise scheduler, and self-conditioning on zero-shot generation. Furthermore, we compare TEncDM with previous approaches"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.19097","kind":"arxiv","version":4},"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/2402.19097/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":"2402.19097","created_at":"2026-07-05T10:18:35.178797+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.19097v4","created_at":"2026-07-05T10:18:35.178797+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.19097","created_at":"2026-07-05T10:18:35.178797+00:00"},{"alias_kind":"pith_short_12","alias_value":"LUNAXDW4RKFA","created_at":"2026-07-05T10:18:35.178797+00:00"},{"alias_kind":"pith_short_16","alias_value":"LUNAXDW4RKFA3QRN","created_at":"2026-07-05T10:18:35.178797+00:00"},{"alias_kind":"pith_short_8","alias_value":"LUNAXDW4","created_at":"2026-07-05T10:18:35.178797+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2505.18853","citing_title":"Smoothie: Smoothing Diffusion on Token Embeddings for Text Generation","ref_index":48,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA","json":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA.json","graph_json":"https://pith.science/api/pith-number/LUNAXDW4RKFA3QRNDJPCEHLLNA/graph.json","events_json":"https://pith.science/api/pith-number/LUNAXDW4RKFA3QRNDJPCEHLLNA/events.json","paper":"https://pith.science/paper/LUNAXDW4"},"agent_actions":{"view_html":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA","download_json":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA.json","view_paper":"https://pith.science/paper/LUNAXDW4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.19097&json=true","fetch_graph":"https://pith.science/api/pith-number/LUNAXDW4RKFA3QRNDJPCEHLLNA/graph.json","fetch_events":"https://pith.science/api/pith-number/LUNAXDW4RKFA3QRNDJPCEHLLNA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA/action/storage_attestation","attest_author":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA/action/author_attestation","sign_citation":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA/action/citation_signature","submit_replication":"https://pith.science/pith/LUNAXDW4RKFA3QRNDJPCEHLLNA/action/replication_record"}},"created_at":"2026-07-05T10:18:35.178797+00:00","updated_at":"2026-07-05T10:18:35.178797+00:00"}