{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:GGAQ7QBWW4IAB4J6LSGROYRUDS","short_pith_number":"pith:GGAQ7QBW","schema_version":"1.0","canonical_sha256":"31810fc036b71000f13e5c8d1762341ca9dbaa9824afc6bdc22ed77604535e11","source":{"kind":"arxiv","id":"2502.20388","version":2},"attestation_state":"computed","paper":{"title":"Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Ju He, Liang-Chieh Chen, Qihang Yu, Sucheng Ren, Xiaohui Shen","submitted_at":"2025-02-27T18:59:08Z","abstract_excerpt":"Autoregressive (AR) modeling, known for its next-token prediction paradigm, underpins state-of-the-art language and visual generative models. Traditionally, a ``token'' is treated as the smallest prediction unit, often a discrete symbol in language or a quantized patch in vision. However, the optimal token definition for 2D image structures remains an open question. Moreover, AR models suffer from exposure bias, where teacher forcing during training leads to error accumulation at inference. In this paper, we propose xAR, a generalized AR framework that extends the notion of a token to an entit"},"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":"2502.20388","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-02-27T18:59:08Z","cross_cats_sorted":[],"title_canon_sha256":"3ca8104159da1e7c88ec8c5d097c87c9d4234f81a3a3ec50bdac3675ec744b8c","abstract_canon_sha256":"ba75580569f6e1ced1bfaf61e236ea18c6d1bbb71ef0a77ec715631e8e38e53c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:36:36.928268Z","signature_b64":"ADhJ0XNyu1mH6coDGtt1dxTF7uf9LBUtLd/6oU5CjtpctYD4SNdOQ3vqQDsJByQsXHoyXu108/2MXmVjQVTUAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31810fc036b71000f13e5c8d1762341ca9dbaa9824afc6bdc22ed77604535e11","last_reissued_at":"2026-07-05T10:36:36.927637Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:36:36.927637Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Next-Token: Next-X Prediction for Autoregressive Visual Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Ju He, Liang-Chieh Chen, Qihang Yu, Sucheng Ren, Xiaohui Shen","submitted_at":"2025-02-27T18:59:08Z","abstract_excerpt":"Autoregressive (AR) modeling, known for its next-token prediction paradigm, underpins state-of-the-art language and visual generative models. Traditionally, a ``token'' is treated as the smallest prediction unit, often a discrete symbol in language or a quantized patch in vision. However, the optimal token definition for 2D image structures remains an open question. Moreover, AR models suffer from exposure bias, where teacher forcing during training leads to error accumulation at inference. In this paper, we propose xAR, a generalized AR framework that extends the notion of a token to an entit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20388","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/2502.20388/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":"2502.20388","created_at":"2026-07-05T10:36:36.927698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20388v2","created_at":"2026-07-05T10:36:36.927698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20388","created_at":"2026-07-05T10:36:36.927698+00:00"},{"alias_kind":"pith_short_12","alias_value":"GGAQ7QBWW4IA","created_at":"2026-07-05T10:36:36.927698+00:00"},{"alias_kind":"pith_short_16","alias_value":"GGAQ7QBWW4IAB4J6","created_at":"2026-07-05T10:36:36.927698+00:00"},{"alias_kind":"pith_short_8","alias_value":"GGAQ7QBW","created_at":"2026-07-05T10:36:36.927698+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26016","citing_title":"MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11096","citing_title":"IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00371","citing_title":"MEPA: Multi-Scale Representation Alignment for Visual Autoregressive Modeling with Mixture of Experts","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26016","citing_title":"MIMFlow: Integrating Masked Image Modeling with Normalizing Flows for End-to-End Image Generation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28386","citing_title":"Data Provenance for Image Auto-Regressive Generation","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30317","citing_title":"VPG: Visual Prefix Guidance for Autoregressive Image and Video Generation","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2602.04883","citing_title":"Protein Autoregressive Modeling via Multiscale Structure Generation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18267","citing_title":"SRC-Flow: Compact Semantic Representations Enable Normalizing Flows for Image Generation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2601.14671","citing_title":"Mirai: Autoregressive Visual Generation Needs Foresight","ref_index":33,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS","json":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS.json","graph_json":"https://pith.science/api/pith-number/GGAQ7QBWW4IAB4J6LSGROYRUDS/graph.json","events_json":"https://pith.science/api/pith-number/GGAQ7QBWW4IAB4J6LSGROYRUDS/events.json","paper":"https://pith.science/paper/GGAQ7QBW"},"agent_actions":{"view_html":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS","download_json":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS.json","view_paper":"https://pith.science/paper/GGAQ7QBW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20388&json=true","fetch_graph":"https://pith.science/api/pith-number/GGAQ7QBWW4IAB4J6LSGROYRUDS/graph.json","fetch_events":"https://pith.science/api/pith-number/GGAQ7QBWW4IAB4J6LSGROYRUDS/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS/action/storage_attestation","attest_author":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS/action/author_attestation","sign_citation":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS/action/citation_signature","submit_replication":"https://pith.science/pith/GGAQ7QBWW4IAB4J6LSGROYRUDS/action/replication_record"}},"created_at":"2026-07-05T10:36:36.927698+00:00","updated_at":"2026-07-05T10:36:36.927698+00:00"}