{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4PUGF33PU3H5NYTS4CBFENSNUU","short_pith_number":"pith:4PUGF33P","schema_version":"1.0","canonical_sha256":"e3e862ef6fa6cfd6e272e08252364da50efa56d6e7054617d995d3681bff5c6f","source":{"kind":"arxiv","id":"2412.02632","version":2},"attestation_state":"computed","paper":{"title":"Scaling Image Tokenizers with Grouped Spherical Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bj\\\"orn Ommer, Jiangtao Wang, Rania Briq, Stefan Kesselheim, Vincent Tao Hu, Yifan Zhang, Zhen Qin","submitted_at":"2024-12-03T18:01:45Z","abstract_excerpt":"Vision tokenizers have gained a lot of attraction due to their scalability and compactness; previous works depend on old-school GAN-based hyperparameters, biased comparisons, and a lack of comprehensive analysis of the scaling behaviours. To tackle those issues, we introduce Grouped Spherical Quantization (GSQ), featuring spherical codebook initialization and lookup regularization to constrain codebook latent to a spherical surface. Our empirical analysis of image tokenizer training strategies demonstrates that GSQ-GAN achieves superior reconstruction quality over state-of-the-art methods with"},"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":"2412.02632","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-12-03T18:01:45Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2e44a2e1a29780d9fbf48caa1d7a71a929a0d083bc3eb0749913341a833833bc","abstract_canon_sha256":"a5f0355c2d0e77df99a4fe82bec7c8290c90e9bd92864f7f0c35fa93bf113682"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:44:18.168525Z","signature_b64":"zA7naKqyP+VEwfTgPfSf1Egd8VBV4SDRYSvVlNVsQfGGH+2q9GNtBorwtYp9OaAXPPY4HcAc1ouSfcrMZ6IOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3e862ef6fa6cfd6e272e08252364da50efa56d6e7054617d995d3681bff5c6f","last_reissued_at":"2026-07-05T09:44:18.168018Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:44:18.168018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Image Tokenizers with Grouped Spherical Quantization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bj\\\"orn Ommer, Jiangtao Wang, Rania Briq, Stefan Kesselheim, Vincent Tao Hu, Yifan Zhang, Zhen Qin","submitted_at":"2024-12-03T18:01:45Z","abstract_excerpt":"Vision tokenizers have gained a lot of attraction due to their scalability and compactness; previous works depend on old-school GAN-based hyperparameters, biased comparisons, and a lack of comprehensive analysis of the scaling behaviours. To tackle those issues, we introduce Grouped Spherical Quantization (GSQ), featuring spherical codebook initialization and lookup regularization to constrain codebook latent to a spherical surface. Our empirical analysis of image tokenizer training strategies demonstrates that GSQ-GAN achieves superior reconstruction quality over state-of-the-art methods with"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.02632","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/2412.02632/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":"2412.02632","created_at":"2026-07-05T09:44:18.168088+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.02632v2","created_at":"2026-07-05T09:44:18.168088+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.02632","created_at":"2026-07-05T09:44:18.168088+00:00"},{"alias_kind":"pith_short_12","alias_value":"4PUGF33PU3H5","created_at":"2026-07-05T09:44:18.168088+00:00"},{"alias_kind":"pith_short_16","alias_value":"4PUGF33PU3H5NYTS","created_at":"2026-07-05T09:44:18.168088+00:00"},{"alias_kind":"pith_short_8","alias_value":"4PUGF33P","created_at":"2026-07-05T09:44:18.168088+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17800","citing_title":"ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU","json":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU.json","graph_json":"https://pith.science/api/pith-number/4PUGF33PU3H5NYTS4CBFENSNUU/graph.json","events_json":"https://pith.science/api/pith-number/4PUGF33PU3H5NYTS4CBFENSNUU/events.json","paper":"https://pith.science/paper/4PUGF33P"},"agent_actions":{"view_html":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU","download_json":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU.json","view_paper":"https://pith.science/paper/4PUGF33P","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.02632&json=true","fetch_graph":"https://pith.science/api/pith-number/4PUGF33PU3H5NYTS4CBFENSNUU/graph.json","fetch_events":"https://pith.science/api/pith-number/4PUGF33PU3H5NYTS4CBFENSNUU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU/action/storage_attestation","attest_author":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU/action/author_attestation","sign_citation":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU/action/citation_signature","submit_replication":"https://pith.science/pith/4PUGF33PU3H5NYTS4CBFENSNUU/action/replication_record"}},"created_at":"2026-07-05T09:44:18.168088+00:00","updated_at":"2026-07-05T09:44:18.168088+00:00"}