{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BIF3BRUFZ4MEPB66IGTLPSBJYW","short_pith_number":"pith:BIF3BRUF","schema_version":"1.0","canonical_sha256":"0a0bb0c685cf184787de41a6b7c829c5a759782d35029583ba9ee9c734779f54","source":{"kind":"arxiv","id":"2606.10199","version":1},"attestation_state":"computed","paper":{"title":"A Continuous-Time Markov Chain Framework for Insertion Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew McCallum, Benjamin Rozonoyer, Dhruvesh Patel, Soumitra Das, Tahira Naseem, Tim G.J. Rudner","submitted_at":"2026-06-08T21:39:43Z","abstract_excerpt":"Insertion Language Models (ILMs) offer several advantages over left-to-right generation and mask-based generation. However, existing formulations of insertion-based generation have largely been ad-hoc. In this paper, we derive a diffusion-style denoising objective for ILMs from first principles by formulating the noising process as a continuous-time Markov chain on the space of variable-length sequences. We show that previous formulations of ILMs can be viewed as special cases of this denoising framework. Through empirical evaluation on a synthetic planning task, we show that the proposed appr"},"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":"2606.10199","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-08T21:39:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"894539d86dc655b454bbecd508a0bb93f1f6e9004671f605f80e92d55fd9bbc0","abstract_canon_sha256":"a6f0c0f8bd3de8c2af6a417ef437f40d93b7310a9d59f001107fe548882a9693"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-10T01:08:59.198375Z","signature_b64":"bDVbEc+9VWSuXQK1w0Ao+wmKWY1luqjDX7K643w1DGsxneuCLrIaRMKc2twsu3pe/cDjbraWkK9npF9klP5IBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a0bb0c685cf184787de41a6b7c829c5a759782d35029583ba9ee9c734779f54","last_reissued_at":"2026-06-10T01:08:59.197387Z","signature_status":"signed_v1","first_computed_at":"2026-06-10T01:08:59.197387Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Continuous-Time Markov Chain Framework for Insertion Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Andrew McCallum, Benjamin Rozonoyer, Dhruvesh Patel, Soumitra Das, Tahira Naseem, Tim G.J. Rudner","submitted_at":"2026-06-08T21:39:43Z","abstract_excerpt":"Insertion Language Models (ILMs) offer several advantages over left-to-right generation and mask-based generation. However, existing formulations of insertion-based generation have largely been ad-hoc. In this paper, we derive a diffusion-style denoising objective for ILMs from first principles by formulating the noising process as a continuous-time Markov chain on the space of variable-length sequences. We show that previous formulations of ILMs can be viewed as special cases of this denoising framework. Through empirical evaluation on a synthetic planning task, we show that the proposed appr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.10199","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/2606.10199/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":"2606.10199","created_at":"2026-06-10T01:08:59.197555+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.10199v1","created_at":"2026-06-10T01:08:59.197555+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.10199","created_at":"2026-06-10T01:08:59.197555+00:00"},{"alias_kind":"pith_short_12","alias_value":"BIF3BRUFZ4ME","created_at":"2026-06-10T01:08:59.197555+00:00"},{"alias_kind":"pith_short_16","alias_value":"BIF3BRUFZ4MEPB66","created_at":"2026-06-10T01:08:59.197555+00:00"},{"alias_kind":"pith_short_8","alias_value":"BIF3BRUF","created_at":"2026-06-10T01:08:59.197555+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/BIF3BRUFZ4MEPB66IGTLPSBJYW","json":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW.json","graph_json":"https://pith.science/api/pith-number/BIF3BRUFZ4MEPB66IGTLPSBJYW/graph.json","events_json":"https://pith.science/api/pith-number/BIF3BRUFZ4MEPB66IGTLPSBJYW/events.json","paper":"https://pith.science/paper/BIF3BRUF"},"agent_actions":{"view_html":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW","download_json":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW.json","view_paper":"https://pith.science/paper/BIF3BRUF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.10199&json=true","fetch_graph":"https://pith.science/api/pith-number/BIF3BRUFZ4MEPB66IGTLPSBJYW/graph.json","fetch_events":"https://pith.science/api/pith-number/BIF3BRUFZ4MEPB66IGTLPSBJYW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW/action/storage_attestation","attest_author":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW/action/author_attestation","sign_citation":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW/action/citation_signature","submit_replication":"https://pith.science/pith/BIF3BRUFZ4MEPB66IGTLPSBJYW/action/replication_record"}},"created_at":"2026-06-10T01:08:59.197555+00:00","updated_at":"2026-06-10T01:08:59.197555+00:00"}