{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:YY7DSZM4RGINERXBR7QDZGCQ6J","short_pith_number":"pith:YY7DSZM4","canonical_record":{"source":{"id":"2305.12223","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-20T16:11:26Z","cross_cats_sorted":[],"title_canon_sha256":"78862a73a0ba2a025d8335f55ff35d5e1fd498d8545973048373c665bb25dd95","abstract_canon_sha256":"72c8e56621cf7888198ac124aa141c6af57a3bc0e1cac1bde6b42cb8efd12627"},"schema_version":"1.0"},"canonical_sha256":"c63e39659c8990d246e18fe03c9850f26913f83d55f9a957d9af900fc0fe0b49","source":{"kind":"arxiv","id":"2305.12223","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.12223","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"arxiv_version","alias_value":"2305.12223v2","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12223","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_12","alias_value":"YY7DSZM4RGIN","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_16","alias_value":"YY7DSZM4RGINERXB","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_8","alias_value":"YY7DSZM4","created_at":"2026-07-05T06:12:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:YY7DSZM4RGINERXBR7QDZGCQ6J","target":"record","payload":{"canonical_record":{"source":{"id":"2305.12223","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-20T16:11:26Z","cross_cats_sorted":[],"title_canon_sha256":"78862a73a0ba2a025d8335f55ff35d5e1fd498d8545973048373c665bb25dd95","abstract_canon_sha256":"72c8e56621cf7888198ac124aa141c6af57a3bc0e1cac1bde6b42cb8efd12627"},"schema_version":"1.0"},"canonical_sha256":"c63e39659c8990d246e18fe03c9850f26913f83d55f9a957d9af900fc0fe0b49","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:12:51.101607Z","signature_b64":"AkaEY0yyp/b2VpY0rZ2ZXlQC+ClSG17Bi3o2Z8WD4+kQ1aevb7KfCYF3UMgdbEOLrpGN925TylgFfqt55QaMAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c63e39659c8990d246e18fe03c9850f26913f83d55f9a957d9af900fc0fe0b49","last_reissued_at":"2026-07-05T06:12:51.101177Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:12:51.101177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2305.12223","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:12:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gkXd5Q1NVeLVFm9cP1Q/Z6EW8hhSTfNqaHeUiTCvz63T/v/NpUyusF8zbXaMsOCkmZ9FQ2EqQDZTdrG0j4LxAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:58:28.160749Z"},"content_sha256":"97ddfc6382a6f8090a8f943175225efe621287f66d5604d3f97d54223f0023ca","schema_version":"1.0","event_id":"sha256:97ddfc6382a6f8090a8f943175225efe621287f66d5604d3f97d54223f0023ca"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:YY7DSZM4RGINERXBR7QDZGCQ6J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Makes for Good Visual Tokenizers for Large Language Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Guangzhi Wang, Mohan Kankanhalli, Xiaohan Ding, Ying Shan, Yixiao Ge","submitted_at":"2023-05-20T16:11:26Z","abstract_excerpt":"We empirically investigate proper pre-training methods to build good visual tokenizers, making Large Language Models (LLMs) powerful Multimodal Large Language Models (MLLMs). In our benchmark, which is curated to evaluate MLLMs visual semantic understanding and fine-grained perception capabilities, we discussed different visual tokenizers pre-trained with dominant methods (i.e., DeiT, CLIP, MAE, DINO), and observe that: i) Fully/weakly supervised models capture more semantics than self-supervised models, but the gap is narrowed by scaling up the pre-training dataset. ii) Self-supervised models"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12223","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/2305.12223/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:12:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wGyo1HDz9SpEet6GOK26K15y5ArznmgzY2ODOz4d7q3PZjG//fXywDb4jF3JNSVA2A0Ze7WMP6L8ADyFrWY3Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T06:58:28.161257Z"},"content_sha256":"4d5938bac70dbb9c901d670472d31aa1a731026f26e9f36fc559def7f8ba1403","schema_version":"1.0","event_id":"sha256:4d5938bac70dbb9c901d670472d31aa1a731026f26e9f36fc559def7f8ba1403"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/bundle.json","state_url":"https://pith.science/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-06T06:58:28Z","links":{"resolver":"https://pith.science/pith/YY7DSZM4RGINERXBR7QDZGCQ6J","bundle":"https://pith.science/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/bundle.json","state":"https://pith.science/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YY7DSZM4RGINERXBR7QDZGCQ6J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:YY7DSZM4RGINERXBR7QDZGCQ6J","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"72c8e56621cf7888198ac124aa141c6af57a3bc0e1cac1bde6b42cb8efd12627","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-20T16:11:26Z","title_canon_sha256":"78862a73a0ba2a025d8335f55ff35d5e1fd498d8545973048373c665bb25dd95"},"schema_version":"1.0","source":{"id":"2305.12223","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.12223","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"arxiv_version","alias_value":"2305.12223v2","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.12223","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_12","alias_value":"YY7DSZM4RGIN","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_16","alias_value":"YY7DSZM4RGINERXB","created_at":"2026-07-05T06:12:51Z"},{"alias_kind":"pith_short_8","alias_value":"YY7DSZM4","created_at":"2026-07-05T06:12:51Z"}],"graph_snapshots":[{"event_id":"sha256:4d5938bac70dbb9c901d670472d31aa1a731026f26e9f36fc559def7f8ba1403","target":"graph","created_at":"2026-07-05T06:12:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2305.12223/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We empirically investigate proper pre-training methods to build good visual tokenizers, making Large Language Models (LLMs) powerful Multimodal Large Language Models (MLLMs). In our benchmark, which is curated to evaluate MLLMs visual semantic understanding and fine-grained perception capabilities, we discussed different visual tokenizers pre-trained with dominant methods (i.e., DeiT, CLIP, MAE, DINO), and observe that: i) Fully/weakly supervised models capture more semantics than self-supervised models, but the gap is narrowed by scaling up the pre-training dataset. ii) Self-supervised models","authors_text":"Guangzhi Wang, Mohan Kankanhalli, Xiaohan Ding, Ying Shan, Yixiao Ge","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-20T16:11:26Z","title":"What Makes for Good Visual Tokenizers for Large Language Models?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.12223","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:97ddfc6382a6f8090a8f943175225efe621287f66d5604d3f97d54223f0023ca","target":"record","created_at":"2026-07-05T06:12:51Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"72c8e56621cf7888198ac124aa141c6af57a3bc0e1cac1bde6b42cb8efd12627","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-20T16:11:26Z","title_canon_sha256":"78862a73a0ba2a025d8335f55ff35d5e1fd498d8545973048373c665bb25dd95"},"schema_version":"1.0","source":{"id":"2305.12223","kind":"arxiv","version":2}},"canonical_sha256":"c63e39659c8990d246e18fe03c9850f26913f83d55f9a957d9af900fc0fe0b49","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c63e39659c8990d246e18fe03c9850f26913f83d55f9a957d9af900fc0fe0b49","first_computed_at":"2026-07-05T06:12:51.101177Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:12:51.101177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"AkaEY0yyp/b2VpY0rZ2ZXlQC+ClSG17Bi3o2Z8WD4+kQ1aevb7KfCYF3UMgdbEOLrpGN925TylgFfqt55QaMAg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:12:51.101607Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.12223","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:97ddfc6382a6f8090a8f943175225efe621287f66d5604d3f97d54223f0023ca","sha256:4d5938bac70dbb9c901d670472d31aa1a731026f26e9f36fc559def7f8ba1403"],"state_sha256":"011e3db7e5513b87d6d05f530c34ef69479321617ab76ff4e663c253ae634e9d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ktlJS37Z6I8WpU6ZN54CM7qfbQdqKutX7g+JSC+A9vQ+g6/Q2ZOUrSJjYOEMO/VDl3KlMP9Yx+eguEedOV2FAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T06:58:28.167012Z","bundle_sha256":"e706ba6b3342e3730cbc0d62bae182b25ab79baf9f900b0eaba7f29ccd061b9f"}}