{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD","short_pith_number":"pith:SEMDPEKU","schema_version":"1.0","canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","source":{"kind":"arxiv","id":"2604.26985","version":2},"attestation_state":"computed","paper":{"title":"Simple Self-Conditioning Adaptation for Masked Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ferdinando Fioretto, Huu Binh Ta, Michael Cardei","submitted_at":"2026-04-28T19:34:04Z","abstract_excerpt":"Masked diffusion models (MDMs) generate discrete sequences by iterative denoising under an absorbing masking process. In standard masked diffusion, if a token remains masked after a reverse update, the model discards its clean-state prediction for that position. Thus, still-masked positions must be repeatedly inferred from the mask token alone. This design choice limits cross-step refinement. To address this limitation, this paper proposes a simple, yet effective, post-training adaptation for MDMs that conditions each denoising step on the model's own previous clean-state predictions. The resu"},"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":"2604.26985","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-28T19:34:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f25407a5fbb1d96396163dc29404036f6006b0d379e78e3e1e6beb4ed054a58b","abstract_canon_sha256":"1e0a09f4cc10b8132d974ba9fb7b554b49bf4fed6f7320b3f480af994e0181aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:05:18.389657Z","signature_b64":"WnzMkf3LKZXWTzrr9wJpLRHXvb2X0hBF0sVfhOTj6k5SFIpnBNDhQcNqiUrSFqJ3cMeOP4kxYggBMaeMnMKPCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","last_reissued_at":"2026-06-09T01:05:18.389213Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:05:18.389213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Simple Self-Conditioning Adaptation for Masked Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Ferdinando Fioretto, Huu Binh Ta, Michael Cardei","submitted_at":"2026-04-28T19:34:04Z","abstract_excerpt":"Masked diffusion models (MDMs) generate discrete sequences by iterative denoising under an absorbing masking process. In standard masked diffusion, if a token remains masked after a reverse update, the model discards its clean-state prediction for that position. Thus, still-masked positions must be repeatedly inferred from the mask token alone. This design choice limits cross-step refinement. To address this limitation, this paper proposes a simple, yet effective, post-training adaptation for MDMs that conditions each denoising step on the model's own previous clean-state predictions. The resu"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SCMDM achieves nearly a 50% reduction in generative perplexity on OWT-trained models (42.89 to 23.72), alongside strong improvements in discretized image synthesis quality, small molecular generation, and enhanced fidelity in genomic distribution modeling.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"Once the model's self-generated clean-state estimates become informative, specialization to refinement is preferable to mixing conditional and unconditional objectives in the post-training regime.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SCMDM adapts trained masked diffusion models to condition denoising steps on their own prior clean predictions, cutting generative perplexity nearly in half on open-web text while improving discretized image, molecule, and genomic synthesis.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1031e3ca8e3a7f6998a969bfdea338dd0f272c0e0d9ddbd8a555a8bbf2969d9f"},"source":{"id":"2604.26985","kind":"arxiv","version":2},"verdict":{"id":"5d41238d-17d8-46ba-b50c-ce4165a35d59","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T16:32:57.468001Z","strongest_claim":"SCMDM achieves nearly a 50% reduction in generative perplexity on OWT-trained models (42.89 to 23.72), alongside strong improvements in discretized image synthesis quality, small molecular generation, and enhanced fidelity in genomic distribution modeling.","one_line_summary":"SCMDM adapts trained masked diffusion models to condition denoising steps on their own prior clean predictions, cutting generative perplexity nearly in half on open-web text while improving discretized image, molecule, and genomic synthesis.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"Once the model's self-generated clean-state estimates become informative, specialization to refinement is preferable to mixing conditional and unconditional objectives in the post-training regime.","pith_extraction_headline":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.26985/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-21T03:35:58.343813Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T20:34:09.529655Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"6c95415a457d647223be2874b32f89b25ad77f7ff86651fd48d783b4597d70e2"},"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":"2604.26985","created_at":"2026-06-09T01:05:18.389273+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.26985v2","created_at":"2026-06-09T01:05:18.389273+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.26985","created_at":"2026-06-09T01:05:18.389273+00:00"},{"alias_kind":"pith_short_12","alias_value":"SEMDPEKUE5WZ","created_at":"2026-06-09T01:05:18.389273+00:00"},{"alias_kind":"pith_short_16","alias_value":"SEMDPEKUE5WZB6HA","created_at":"2026-06-09T01:05:18.389273+00:00"},{"alias_kind":"pith_short_8","alias_value":"SEMDPEKU","created_at":"2026-06-09T01:05:18.389273+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.00588","citing_title":"Low Perplexity is Repetition: A One-Dimensional Self-Conditioning Attractor in Continuous Diffusion LMs","ref_index":1,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD","json":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD.json","graph_json":"https://pith.science/api/pith-number/SEMDPEKUE5WZB6HAAWJMOI22AD/graph.json","events_json":"https://pith.science/api/pith-number/SEMDPEKUE5WZB6HAAWJMOI22AD/events.json","paper":"https://pith.science/paper/SEMDPEKU"},"agent_actions":{"view_html":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD","download_json":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD.json","view_paper":"https://pith.science/paper/SEMDPEKU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.26985&json=true","fetch_graph":"https://pith.science/api/pith-number/SEMDPEKUE5WZB6HAAWJMOI22AD/graph.json","fetch_events":"https://pith.science/api/pith-number/SEMDPEKUE5WZB6HAAWJMOI22AD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/action/storage_attestation","attest_author":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/action/author_attestation","sign_citation":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/action/citation_signature","submit_replication":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/action/replication_record"}},"created_at":"2026-06-09T01:05:18.389273+00:00","updated_at":"2026-06-09T01:05:18.389273+00:00"}