{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6EZGBKXBE5B7BNPXLUP3AQ22FI","short_pith_number":"pith:6EZGBKXB","schema_version":"1.0","canonical_sha256":"f13260aae12743f0b5f75d1fb0435a2a1ae95d38fbc3bfa63f8c2e956f4f0da1","source":{"kind":"arxiv","id":"2606.02133","version":1},"attestation_state":"computed","paper":{"title":"Variational Learning for Insertion-based Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arthur Gretton, David van Dijk, Jiaxin Shi, Michalis K. Titsias, Rex Ying, Yangtian Zhang, Zhe Wang","submitted_at":"2026-06-01T11:59:46Z","abstract_excerpt":"Non-monotonic sequence generation methods, such as masked diffusion models, provide a flexible alternative to left-to-right autoregressive modeling by allowing tokens to be generated in non-fixed and prescribed orders. Despite their practical advantages, most existing non-monotonic models are order-agnostic and rely on a fixed-length grid, limiting their ability to support variable-length generation and adaptive insertion order. In this work, we introduce a probabilistic framework for learning insertion order in variable-length insertion models. We formalize a bijective correspondence between "},"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":"2606.02133","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-01T11:59:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a6349ad7b379546b5843ee58dc08362a9da57525ef6a68c56f14df2588404d81","abstract_canon_sha256":"6737ee15d395437e94984ac21574fc81b01e5cc57702c401e6a5c878e6f596c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:05:07.394900Z","signature_b64":"IKWPC41LtgEbdr5qcXQT45t/yp1GI9DE6XiYPVHwOIQdIZ4zk+z6u5fOh5aDTBWidEyC5wjoFg/fqEC+OZUuDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f13260aae12743f0b5f75d1fb0435a2a1ae95d38fbc3bfa63f8c2e956f4f0da1","last_reissued_at":"2026-06-02T02:05:07.394486Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:05:07.394486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Variational Learning for Insertion-based Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Arthur Gretton, David van Dijk, Jiaxin Shi, Michalis K. Titsias, Rex Ying, Yangtian Zhang, Zhe Wang","submitted_at":"2026-06-01T11:59:46Z","abstract_excerpt":"Non-monotonic sequence generation methods, such as masked diffusion models, provide a flexible alternative to left-to-right autoregressive modeling by allowing tokens to be generated in non-fixed and prescribed orders. Despite their practical advantages, most existing non-monotonic models are order-agnostic and rely on a fixed-length grid, limiting their ability to support variable-length generation and adaptive insertion order. In this work, we introduce a probabilistic framework for learning insertion order in variable-length insertion models. We formalize a bijective correspondence between "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02133","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/2606.02133/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":"2606.02133","created_at":"2026-06-02T02:05:07.394547+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02133v1","created_at":"2026-06-02T02:05:07.394547+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02133","created_at":"2026-06-02T02:05:07.394547+00:00"},{"alias_kind":"pith_short_12","alias_value":"6EZGBKXBE5B7","created_at":"2026-06-02T02:05:07.394547+00:00"},{"alias_kind":"pith_short_16","alias_value":"6EZGBKXBE5B7BNPX","created_at":"2026-06-02T02:05:07.394547+00:00"},{"alias_kind":"pith_short_8","alias_value":"6EZGBKXB","created_at":"2026-06-02T02:05:07.394547+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/6EZGBKXBE5B7BNPXLUP3AQ22FI","json":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI.json","graph_json":"https://pith.science/api/pith-number/6EZGBKXBE5B7BNPXLUP3AQ22FI/graph.json","events_json":"https://pith.science/api/pith-number/6EZGBKXBE5B7BNPXLUP3AQ22FI/events.json","paper":"https://pith.science/paper/6EZGBKXB"},"agent_actions":{"view_html":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI","download_json":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI.json","view_paper":"https://pith.science/paper/6EZGBKXB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02133&json=true","fetch_graph":"https://pith.science/api/pith-number/6EZGBKXBE5B7BNPXLUP3AQ22FI/graph.json","fetch_events":"https://pith.science/api/pith-number/6EZGBKXBE5B7BNPXLUP3AQ22FI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI/action/storage_attestation","attest_author":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI/action/author_attestation","sign_citation":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI/action/citation_signature","submit_replication":"https://pith.science/pith/6EZGBKXBE5B7BNPXLUP3AQ22FI/action/replication_record"}},"created_at":"2026-06-02T02:05:07.394547+00:00","updated_at":"2026-06-02T02:05:07.394547+00:00"}