{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T6ZXFIM2EBAPOGNYIIKSNLEN5Y","short_pith_number":"pith:T6ZXFIM2","schema_version":"1.0","canonical_sha256":"9fb372a19a2040f719b8421526ac8dee287f659509452485ca04612b97faf133","source":{"kind":"arxiv","id":"2606.32039","version":1},"attestation_state":"computed","paper":{"title":"GEAR: Guided End-to-End AutoRegression for Image Synthesis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Chenguo Lin, Jianwei Zhang, Liefeng Bo, Li Yuan, Miles Yang, Sixiang Chen, Yunlong Lin, Yunyang Ge, Zhao Zhong, Zheyuan Liu","submitted_at":"2026-06-30T17:59:57Z","abstract_excerpt":"Visual generative models are typically trained in two stages. A tokenizer is first trained for reconstruction and then frozen, after which a generator is trained on its discrete indices or continuous latents. This decoupling leaves the tokenizer unaware of what the generator finds easy to model. We present GEAR (Guided End-to-end AutoRegression), which trains a vector-quantized (VQ) tokenizer and an autoregressive (AR) generator jointly and end-to-end, guided by representation alignment. The key obstacle is that the VQ index fed to the AR model is non-differentiable, so gradients cannot reach "},"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":"2606.32039","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-30T17:59:57Z","cross_cats_sorted":[],"title_canon_sha256":"9d99d435769758d6131a2c26c398cedf48a98b41bf404c4b984ee3bb317f3faf","abstract_canon_sha256":"4d004026551c6e5ba062b571533c87d6d7eb60325c7480931c66ee02e916cced"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-01T02:17:48.057198Z","signature_b64":"doaEdOD5IooTTrmfniELKpTCmceCm/BxQjnEV/vQyOStJe+9IxCHrk3KR5oqwRcU8fhLDtE/OG6Frz53YCWYDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9fb372a19a2040f719b8421526ac8dee287f659509452485ca04612b97faf133","last_reissued_at":"2026-07-01T02:17:48.056716Z","signature_status":"signed_v1","first_computed_at":"2026-07-01T02:17:48.056716Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GEAR: Guided End-to-End AutoRegression for Image Synthesis","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bin Lin, Chenguo Lin, Jianwei Zhang, Liefeng Bo, Li Yuan, Miles Yang, Sixiang Chen, Yunlong Lin, Yunyang Ge, Zhao Zhong, Zheyuan Liu","submitted_at":"2026-06-30T17:59:57Z","abstract_excerpt":"Visual generative models are typically trained in two stages. A tokenizer is first trained for reconstruction and then frozen, after which a generator is trained on its discrete indices or continuous latents. This decoupling leaves the tokenizer unaware of what the generator finds easy to model. We present GEAR (Guided End-to-end AutoRegression), which trains a vector-quantized (VQ) tokenizer and an autoregressive (AR) generator jointly and end-to-end, guided by representation alignment. The key obstacle is that the VQ index fed to the AR model is non-differentiable, so gradients cannot reach "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.32039","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/2606.32039/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":"2606.32039","created_at":"2026-07-01T02:17:48.056781+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.32039v1","created_at":"2026-07-01T02:17:48.056781+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.32039","created_at":"2026-07-01T02:17:48.056781+00:00"},{"alias_kind":"pith_short_12","alias_value":"T6ZXFIM2EBAP","created_at":"2026-07-01T02:17:48.056781+00:00"},{"alias_kind":"pith_short_16","alias_value":"T6ZXFIM2EBAPOGNY","created_at":"2026-07-01T02:17:48.056781+00:00"},{"alias_kind":"pith_short_8","alias_value":"T6ZXFIM2","created_at":"2026-07-01T02:17:48.056781+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/T6ZXFIM2EBAPOGNYIIKSNLEN5Y","json":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y.json","graph_json":"https://pith.science/api/pith-number/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/graph.json","events_json":"https://pith.science/api/pith-number/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/events.json","paper":"https://pith.science/paper/T6ZXFIM2"},"agent_actions":{"view_html":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y","download_json":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y.json","view_paper":"https://pith.science/paper/T6ZXFIM2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.32039&json=true","fetch_graph":"https://pith.science/api/pith-number/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/graph.json","fetch_events":"https://pith.science/api/pith-number/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/action/storage_attestation","attest_author":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/action/author_attestation","sign_citation":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/action/citation_signature","submit_replication":"https://pith.science/pith/T6ZXFIM2EBAPOGNYIIKSNLEN5Y/action/replication_record"}},"created_at":"2026-07-01T02:17:48.056781+00:00","updated_at":"2026-07-01T02:17:48.056781+00:00"}