{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:I5LCFXQV6MNT4ZNQZVWGLZO2YV","short_pith_number":"pith:I5LCFXQV","schema_version":"1.0","canonical_sha256":"475622de15f31b3e65b0cd6c65e5dac5643cb394dd6cb532be6861779d354753","source":{"kind":"arxiv","id":"2509.01025","version":2},"attestation_state":"computed","paper":{"title":"Any-Order Flexible Length Masked Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Carles Domingo-Enrich, Jaeyeon Kim, Lee Cheuk-Kit, Michael Albergo, Sham Kakade, Sitan Chen, Timothy Ngotiaoco, Yilun Du","submitted_at":"2025-08-31T23:34:53Z","abstract_excerpt":"Masked diffusion models (MDMs) have recently emerged as a promising alternative to autoregressive models over discrete domains. MDMs generate sequences in an any-order, parallel fashion, enabling fast inference and strong performance on non-causal tasks. However, a crucial limitation is that they do not support token insertions and are thus limited to fixed-length generations. To this end, we introduce Flexible Masked Diffusion Models (FlexMDMs), a discrete diffusion paradigm that simultaneously can model sequences of flexible length while provably retaining MDMs' flexibility of any-order infe"},"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":"2509.01025","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-08-31T23:34:53Z","cross_cats_sorted":[],"title_canon_sha256":"a2838cee46d018791b963c7cd670257601a7f37d71f5ccbb3d007868ced885fb","abstract_canon_sha256":"ca5e88ad66ef991834170480eb16c2bdf200728006f261f259f92552c7fc25f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:19.437078Z","signature_b64":"L+16HcHW/v/0OVv/RaP13geOobf//zs5Qa62w0KQvspSYJ4WRBqV0kSbGiWOB/0bZ7w68XEU9SX1qFoa/4ypAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"475622de15f31b3e65b0cd6c65e5dac5643cb394dd6cb532be6861779d354753","last_reissued_at":"2026-07-05T12:06:19.436581Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:19.436581Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Any-Order Flexible Length Masked Diffusion","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Carles Domingo-Enrich, Jaeyeon Kim, Lee Cheuk-Kit, Michael Albergo, Sham Kakade, Sitan Chen, Timothy Ngotiaoco, Yilun Du","submitted_at":"2025-08-31T23:34:53Z","abstract_excerpt":"Masked diffusion models (MDMs) have recently emerged as a promising alternative to autoregressive models over discrete domains. MDMs generate sequences in an any-order, parallel fashion, enabling fast inference and strong performance on non-causal tasks. However, a crucial limitation is that they do not support token insertions and are thus limited to fixed-length generations. To this end, we introduce Flexible Masked Diffusion Models (FlexMDMs), a discrete diffusion paradigm that simultaneously can model sequences of flexible length while provably retaining MDMs' flexibility of any-order infe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01025","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/2509.01025/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":"2509.01025","created_at":"2026-07-05T12:06:19.436636+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.01025v2","created_at":"2026-07-05T12:06:19.436636+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01025","created_at":"2026-07-05T12:06:19.436636+00:00"},{"alias_kind":"pith_short_12","alias_value":"I5LCFXQV6MNT","created_at":"2026-07-05T12:06:19.436636+00:00"},{"alias_kind":"pith_short_16","alias_value":"I5LCFXQV6MNT4ZNQ","created_at":"2026-07-05T12:06:19.436636+00:00"},{"alias_kind":"pith_short_8","alias_value":"I5LCFXQV","created_at":"2026-07-05T12:06:19.436636+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17999","citing_title":"VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01775","citing_title":"Set Diffusion: Interpolating Token Orderings Between Autoregression and Diffusion for Fast and Flexible Decoding","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07239","citing_title":"Generative Molecular Morphing for Flexible-Size Design via Unbalanced Optimal Transport","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02955","citing_title":"Fast-dLLM++: Fr\\'{e}chet Profile Decoding for Faster Diffusion LLM Inference","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2510.06133","citing_title":"CreditDecoding: Accelerating Parallel Decoding in Diffusion Large Language Models with Trace Credit","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2512.10857","citing_title":"Generative Modeling from Black-box Corruptions via Self-Consistent Stochastic Interpolants","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09603","citing_title":"Edit-Based Refinement for Parallel Masked Diffusion Language Models","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18739","citing_title":"Discrete Tilt Matching","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV","json":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV.json","graph_json":"https://pith.science/api/pith-number/I5LCFXQV6MNT4ZNQZVWGLZO2YV/graph.json","events_json":"https://pith.science/api/pith-number/I5LCFXQV6MNT4ZNQZVWGLZO2YV/events.json","paper":"https://pith.science/paper/I5LCFXQV"},"agent_actions":{"view_html":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV","download_json":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV.json","view_paper":"https://pith.science/paper/I5LCFXQV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.01025&json=true","fetch_graph":"https://pith.science/api/pith-number/I5LCFXQV6MNT4ZNQZVWGLZO2YV/graph.json","fetch_events":"https://pith.science/api/pith-number/I5LCFXQV6MNT4ZNQZVWGLZO2YV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV/action/storage_attestation","attest_author":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV/action/author_attestation","sign_citation":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV/action/citation_signature","submit_replication":"https://pith.science/pith/I5LCFXQV6MNT4ZNQZVWGLZO2YV/action/replication_record"}},"created_at":"2026-07-05T12:06:19.436636+00:00","updated_at":"2026-07-05T12:06:19.436636+00:00"}