{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Z6LCDAHZ4UEO6F7OAVXKLTSIVH","short_pith_number":"pith:Z6LCDAHZ","schema_version":"1.0","canonical_sha256":"cf962180f9e508ef17ee056ea5ce48a9cf5f66008b19db21811f68fbb613f070","source":{"kind":"arxiv","id":"2605.31215","version":1},"attestation_state":"computed","paper":{"title":"Fixed-Point Masked Generative Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Alba Carballo-Castro, Andrea Miele, Justin Deschenaux, Pascal Frossard, Yiming Qin","submitted_at":"2026-05-29T12:19:26Z","abstract_excerpt":"Masked Generative Models (MGMs) enable parallel decoding and achieve strong performance across modalities, but require full-sequence bidirectional transformers at every step, making training costly and degrading quality under low sampling budgets. Existing work improves efficiency via better samplers or cheaper fixed-depth denoisers, but they still allocate a fixed amount of denoiser computation to each refinement step. We introduce Fixed-Point Masked Generative Models (FP-MGMs), which replace part of the denoiser with a fixed-point solver over shared attention layers to enable adaptive depth "},"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":"2605.31215","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-29T12:19:26Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"e25f8c35df14fd1e9fe0c7305696f35f8ffa823eda6651ef39c3871662600292","abstract_canon_sha256":"ae193866bd9174b544c71ae880e28e2db0c853d09a1815e9754639b547934d54"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:04:04.278012Z","signature_b64":"p8w32rC5Wr1fUUeRfu5R3+IQATXNyV5pPEN1qM6vYYHI0acRIYKbyfZLUq1RsawvE790vg8/BaObc1QQLztaAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cf962180f9e508ef17ee056ea5ce48a9cf5f66008b19db21811f68fbb613f070","last_reissued_at":"2026-06-01T01:04:04.277442Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:04:04.277442Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fixed-Point Masked Generative Modeling","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Alba Carballo-Castro, Andrea Miele, Justin Deschenaux, Pascal Frossard, Yiming Qin","submitted_at":"2026-05-29T12:19:26Z","abstract_excerpt":"Masked Generative Models (MGMs) enable parallel decoding and achieve strong performance across modalities, but require full-sequence bidirectional transformers at every step, making training costly and degrading quality under low sampling budgets. Existing work improves efficiency via better samplers or cheaper fixed-depth denoisers, but they still allocate a fixed amount of denoiser computation to each refinement step. We introduce Fixed-Point Masked Generative Models (FP-MGMs), which replace part of the denoiser with a fixed-point solver over shared attention layers to enable adaptive depth "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.31215","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/2605.31215/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":"2605.31215","created_at":"2026-06-01T01:04:04.277535+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.31215v1","created_at":"2026-06-01T01:04:04.277535+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.31215","created_at":"2026-06-01T01:04:04.277535+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z6LCDAHZ4UEO","created_at":"2026-06-01T01:04:04.277535+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z6LCDAHZ4UEO6F7O","created_at":"2026-06-01T01:04:04.277535+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z6LCDAHZ","created_at":"2026-06-01T01:04:04.277535+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/Z6LCDAHZ4UEO6F7OAVXKLTSIVH","json":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH.json","graph_json":"https://pith.science/api/pith-number/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/graph.json","events_json":"https://pith.science/api/pith-number/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/events.json","paper":"https://pith.science/paper/Z6LCDAHZ"},"agent_actions":{"view_html":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH","download_json":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH.json","view_paper":"https://pith.science/paper/Z6LCDAHZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.31215&json=true","fetch_graph":"https://pith.science/api/pith-number/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/graph.json","fetch_events":"https://pith.science/api/pith-number/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/action/storage_attestation","attest_author":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/action/author_attestation","sign_citation":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/action/citation_signature","submit_replication":"https://pith.science/pith/Z6LCDAHZ4UEO6F7OAVXKLTSIVH/action/replication_record"}},"created_at":"2026-06-01T01:04:04.277535+00:00","updated_at":"2026-06-01T01:04:04.277535+00:00"}