{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD","short_pith_number":"pith:SEMDPEKU","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"},"canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","source":{"kind":"arxiv","id":"2604.26985","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.26985","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"arxiv_version","alias_value":"2604.26985v2","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.26985","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_12","alias_value":"SEMDPEKUE5WZ","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_16","alias_value":"SEMDPEKUE5WZB6HA","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_8","alias_value":"SEMDPEKU","created_at":"2026-06-09T01:05:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD","target":"record","payload":{"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"},"canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","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"},"source_kind":"arxiv","source_id":"2604.26985","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-09T01:05:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"N8SH26I9epyjvUZAB8YDEOOuJaHIny6AKy9kpS0+sQOsVfqbuoPHOBNFZL6t++SKj+VkxbWgcqNaYaGrTT9ECA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T11:08:36.127366Z"},"content_sha256":"b9f867f119c2d36a9d03812e85ea451a6e7cd4b323cd6b83a255dae2db490365","schema_version":"1.0","event_id":"sha256:b9f867f119c2d36a9d03812e85ea451a6e7cd4b323cd6b83a255dae2db490365"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD","target":"graph","payload":{"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"},"verdict_id":"5d41238d-17d8-46ba-b50c-ce4165a35d59"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-09T01:05:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MEBmVz3Qra8TB1kRi8T96ilgtjdIwmsVmbVa/6zPQJ5Jf6324DF4h4TlKhItazN5USYg9RT9WDqnoRLStmcKCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-03T11:08:36.127882Z"},"content_sha256":"de2b0b4593f23fd1a6a594afae55e11b4ab194dba81a3b3238540cc436d2489d","schema_version":"1.0","event_id":"sha256:de2b0b4593f23fd1a6a594afae55e11b4ab194dba81a3b3238540cc436d2489d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/bundle.json","state_url":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-03T11:08:36Z","links":{"resolver":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD","bundle":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/bundle.json","state":"https://pith.science/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SEMDPEKUE5WZB6HAAWJMOI22AD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SEMDPEKUE5WZB6HAAWJMOI22AD","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"1e0a09f4cc10b8132d974ba9fb7b554b49bf4fed6f7320b3f480af994e0181aa","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-28T19:34:04Z","title_canon_sha256":"f25407a5fbb1d96396163dc29404036f6006b0d379e78e3e1e6beb4ed054a58b"},"schema_version":"1.0","source":{"id":"2604.26985","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.26985","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"arxiv_version","alias_value":"2604.26985v2","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.26985","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_12","alias_value":"SEMDPEKUE5WZ","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_16","alias_value":"SEMDPEKUE5WZB6HA","created_at":"2026-06-09T01:05:18Z"},{"alias_kind":"pith_short_8","alias_value":"SEMDPEKU","created_at":"2026-06-09T01:05:18Z"}],"graph_snapshots":[{"event_id":"sha256:de2b0b4593f23fd1a6a594afae55e11b4ab194dba81a3b3238540cc436d2489d","target":"graph","created_at":"2026-06-09T01:05:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps."}],"snapshot_sha256":"1031e3ca8e3a7f6998a969bfdea338dd0f272c0e0d9ddbd8a555a8bbf2969d9f"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2604.26985/integrity.json","findings":[],"snapshot_sha256":"6c95415a457d647223be2874b32f89b25ad77f7ff86651fd48d783b4597d70e2","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Ferdinando Fioretto, Huu Binh Ta, Michael Cardei","cross_cats":["cs.AI"],"headline":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-28T19:34:04Z","title":"Simple Self-Conditioning Adaptation for Masked Diffusion Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.26985","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-07T16:32:57.468001Z","id":"5d41238d-17d8-46ba-b50c-ce4165a35d59","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"A post-training adaptation conditions masked diffusion models on their own prior clean predictions to enable repeated refinement across denoising steps.","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.","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."}},"verdict_id":"5d41238d-17d8-46ba-b50c-ce4165a35d59"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b9f867f119c2d36a9d03812e85ea451a6e7cd4b323cd6b83a255dae2db490365","target":"record","created_at":"2026-06-09T01:05:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"1e0a09f4cc10b8132d974ba9fb7b554b49bf4fed6f7320b3f480af994e0181aa","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-04-28T19:34:04Z","title_canon_sha256":"f25407a5fbb1d96396163dc29404036f6006b0d379e78e3e1e6beb4ed054a58b"},"schema_version":"1.0","source":{"id":"2604.26985","kind":"arxiv","version":2}},"canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9118379154276d90f8e00592c7235a00ee279d68d87bf4eb609a3b1a3e1e44aa","first_computed_at":"2026-06-09T01:05:18.389213Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T01:05:18.389213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WnzMkf3LKZXWTzrr9wJpLRHXvb2X0hBF0sVfhOTj6k5SFIpnBNDhQcNqiUrSFqJ3cMeOP4kxYggBMaeMnMKPCQ==","signature_status":"signed_v1","signed_at":"2026-06-09T01:05:18.389657Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.26985","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b9f867f119c2d36a9d03812e85ea451a6e7cd4b323cd6b83a255dae2db490365","sha256:de2b0b4593f23fd1a6a594afae55e11b4ab194dba81a3b3238540cc436d2489d"],"state_sha256":"9daba15d6d9a46bd8458e133d2218ffe3698071afaa4e8d1345c6533a481f8e4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fymN/yKpRfcFW9b3UwNF/h7dLhyBA/M5K5JqM6PgFC2O+aLiaAvjpGQTn4JbHRvY90TOFCO/ZZHqqLb6EwgXDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-03T11:08:36.130224Z","bundle_sha256":"966a2503a5fc8a563f74857cfb4c64459b90ea846043e35900b1c03812fd8803"}}