{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TRQ4K3GNRNE2JLVX6FPR7N2AWU","short_pith_number":"pith:TRQ4K3GN","schema_version":"1.0","canonical_sha256":"9c61c56ccd8b49a4aeb7f15f1fb740b52ba9808bbbe576e011c3a40290b1eb17","source":{"kind":"arxiv","id":"2412.01827","version":2},"attestation_state":"computed","paper":{"title":"RandAR: Decoder-only Autoregressive Visual Generation in Random Orders","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Fujun Luan, Hao Tan, Kai Zhang, Tianyuan Zhang, William T. Freeman, Yunze Man, Yu-Xiong Wang, Ziqi Pang","submitted_at":"2024-12-02T18:59:53Z","abstract_excerpt":"We introduce RandAR, a decoder-only visual autoregressive (AR) model capable of generating images in arbitrary token orders. Unlike previous decoder-only AR models that rely on a predefined generation order, RandAR removes this inductive bias, unlocking new capabilities in decoder-only generation. Our essential design enables random order by inserting a \"position instruction token\" before each image token to be predicted, representing the spatial location of the next image token. Trained on randomly permuted token sequences -- a more challenging task than fixed-order generation, RandAR achieve"},"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":"2412.01827","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-02T18:59:53Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5d85bcce3d2e752121fa86c6951d8b788163fa70a7e1f876aff8033478c58693","abstract_canon_sha256":"3c6466434414a11a2afad4ca654e742fefa6afe9900dcfd2e5652ed7bbb8210e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:18.068511Z","signature_b64":"SpmiCRyPtxeYp7aPUGiV0jq8ZVrsngRAKcFf1g3NL+c8P4jCiQ+EdMQbiMF9CBsRby5FY3t/61IytdSgkK45BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c61c56ccd8b49a4aeb7f15f1fb740b52ba9808bbbe576e011c3a40290b1eb17","last_reissued_at":"2026-07-05T11:33:18.068091Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:18.068091Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RandAR: Decoder-only Autoregressive Visual Generation in Random Orders","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Fujun Luan, Hao Tan, Kai Zhang, Tianyuan Zhang, William T. Freeman, Yunze Man, Yu-Xiong Wang, Ziqi Pang","submitted_at":"2024-12-02T18:59:53Z","abstract_excerpt":"We introduce RandAR, a decoder-only visual autoregressive (AR) model capable of generating images in arbitrary token orders. Unlike previous decoder-only AR models that rely on a predefined generation order, RandAR removes this inductive bias, unlocking new capabilities in decoder-only generation. Our essential design enables random order by inserting a \"position instruction token\" before each image token to be predicted, representing the spatial location of the next image token. Trained on randomly permuted token sequences -- a more challenging task than fixed-order generation, RandAR achieve"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01827","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/2412.01827/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":"2412.01827","created_at":"2026-07-05T11:33:18.068147+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.01827v2","created_at":"2026-07-05T11:33:18.068147+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01827","created_at":"2026-07-05T11:33:18.068147+00:00"},{"alias_kind":"pith_short_12","alias_value":"TRQ4K3GNRNE2","created_at":"2026-07-05T11:33:18.068147+00:00"},{"alias_kind":"pith_short_16","alias_value":"TRQ4K3GNRNE2JLVX","created_at":"2026-07-05T11:33:18.068147+00:00"},{"alias_kind":"pith_short_8","alias_value":"TRQ4K3GN","created_at":"2026-07-05T11:33:18.068147+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.15564","citing_title":"Show-o2: Improved Native Unified Multimodal Models","ref_index":84,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU","json":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU.json","graph_json":"https://pith.science/api/pith-number/TRQ4K3GNRNE2JLVX6FPR7N2AWU/graph.json","events_json":"https://pith.science/api/pith-number/TRQ4K3GNRNE2JLVX6FPR7N2AWU/events.json","paper":"https://pith.science/paper/TRQ4K3GN"},"agent_actions":{"view_html":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU","download_json":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU.json","view_paper":"https://pith.science/paper/TRQ4K3GN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.01827&json=true","fetch_graph":"https://pith.science/api/pith-number/TRQ4K3GNRNE2JLVX6FPR7N2AWU/graph.json","fetch_events":"https://pith.science/api/pith-number/TRQ4K3GNRNE2JLVX6FPR7N2AWU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU/action/storage_attestation","attest_author":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU/action/author_attestation","sign_citation":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU/action/citation_signature","submit_replication":"https://pith.science/pith/TRQ4K3GNRNE2JLVX6FPR7N2AWU/action/replication_record"}},"created_at":"2026-07-05T11:33:18.068147+00:00","updated_at":"2026-07-05T11:33:18.068147+00:00"}