{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:UWAX6BIVTSCA7D6RK3ITKYCKYP","short_pith_number":"pith:UWAX6BIV","canonical_record":{"source":{"id":"2401.09865","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-18T10:28:45Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c44c46b5de2c29aa43cc9271d8f32e261eedf29451733038d5fda5bce1fed2ba","abstract_canon_sha256":"b76d98e4e8d6d957f61600ebc992ca60932aebc5a662499ddf68dc1545aa908e"},"schema_version":"1.0"},"canonical_sha256":"a5817f05159c840f8fd156d135604ac3e7d9604589cd4eff6c419db22236b437","source":{"kind":"arxiv","id":"2401.09865","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.09865","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"arxiv_version","alias_value":"2401.09865v1","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.09865","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_12","alias_value":"UWAX6BIVTSCA","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_16","alias_value":"UWAX6BIVTSCA7D6R","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_8","alias_value":"UWAX6BIV","created_at":"2026-07-05T07:35:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:UWAX6BIVTSCA7D6RK3ITKYCKYP","target":"record","payload":{"canonical_record":{"source":{"id":"2401.09865","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-18T10:28:45Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c44c46b5de2c29aa43cc9271d8f32e261eedf29451733038d5fda5bce1fed2ba","abstract_canon_sha256":"b76d98e4e8d6d957f61600ebc992ca60932aebc5a662499ddf68dc1545aa908e"},"schema_version":"1.0"},"canonical_sha256":"a5817f05159c840f8fd156d135604ac3e7d9604589cd4eff6c419db22236b437","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:35:03.492645Z","signature_b64":"8AYG7KdFAAsOSCoimf0aO4crrrEfNqFQljgJqXCp8FHph4B29pEAmgvnQ96QX9lmvYbSWxmLbIS031pJv3ReAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5817f05159c840f8fd156d135604ac3e7d9604589cd4eff6c419db22236b437","last_reissued_at":"2026-07-05T07:35:03.492211Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:35:03.492211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.09865","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:35:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1xXQgUstyt/KUqV2cQrSLOjcc/wdGkvNqxOkZYRJ5h3CURO0FlcCweX95fSWDN14dvbck4cdMJfuuIlS0wzOAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T20:21:03.394077Z"},"content_sha256":"ede75be3e20cf5ca0a0ff84932459338aee95f69aad68738009a6c806782b515","schema_version":"1.0","event_id":"sha256:ede75be3e20cf5ca0a0ff84932459338aee95f69aad68738009a6c806782b515"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:UWAX6BIVTSCA7D6RK3ITKYCKYP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving fine-grained understanding in image-text pre-training","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexey A. Gritsenko, Anastasija Ili\\'c, Charles Blundell, Christos Kaplanis, Goker Erdogan, Ioana Bica, Jovana Mitrovi\\'c, Matko Bo\\v{s}njak, Matthias Bauer, Matthias Minderer, Razvan Pascanu","submitted_at":"2024-01-18T10:28:45Z","abstract_excerpt":"We introduce SPARse Fine-grained Contrastive Alignment (SPARC), a simple method for pretraining more fine-grained multimodal representations from image-text pairs. Given that multiple image patches often correspond to single words, we propose to learn a grouping of image patches for every token in the caption. To achieve this, we use a sparse similarity metric between image patches and language tokens and compute for each token a language-grouped vision embedding as the weighted average of patches. The token and language-grouped vision embeddings are then contrasted through a fine-grained sequ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.09865","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/2401.09865/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":"2026-07-05T07:35:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bIHFFrhV33xeKy+3tjb4R0GWGS4aZTAEGklg+h25rov9In3ooSLdkt59gn0pSPzqQWscfrERrQJbCEjl7wPdAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T20:21:03.394592Z"},"content_sha256":"4372bae660d5498aa6d64cc8e613161ac074de508a212f9fb51f4c25757ba9c3","schema_version":"1.0","event_id":"sha256:4372bae660d5498aa6d64cc8e613161ac074de508a212f9fb51f4c25757ba9c3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/bundle.json","state_url":"https://pith.science/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/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":"2026-08-03T20:21:03Z","links":{"resolver":"https://pith.science/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP","bundle":"https://pith.science/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/bundle.json","state":"https://pith.science/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UWAX6BIVTSCA7D6RK3ITKYCKYP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:UWAX6BIVTSCA7D6RK3ITKYCKYP","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":"b76d98e4e8d6d957f61600ebc992ca60932aebc5a662499ddf68dc1545aa908e","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-18T10:28:45Z","title_canon_sha256":"c44c46b5de2c29aa43cc9271d8f32e261eedf29451733038d5fda5bce1fed2ba"},"schema_version":"1.0","source":{"id":"2401.09865","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.09865","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"arxiv_version","alias_value":"2401.09865v1","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.09865","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_12","alias_value":"UWAX6BIVTSCA","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_16","alias_value":"UWAX6BIVTSCA7D6R","created_at":"2026-07-05T07:35:03Z"},{"alias_kind":"pith_short_8","alias_value":"UWAX6BIV","created_at":"2026-07-05T07:35:03Z"}],"graph_snapshots":[{"event_id":"sha256:4372bae660d5498aa6d64cc8e613161ac074de508a212f9fb51f4c25757ba9c3","target":"graph","created_at":"2026-07-05T07:35:03Z","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/2401.09865/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce SPARse Fine-grained Contrastive Alignment (SPARC), a simple method for pretraining more fine-grained multimodal representations from image-text pairs. Given that multiple image patches often correspond to single words, we propose to learn a grouping of image patches for every token in the caption. To achieve this, we use a sparse similarity metric between image patches and language tokens and compute for each token a language-grouped vision embedding as the weighted average of patches. The token and language-grouped vision embeddings are then contrasted through a fine-grained sequ","authors_text":"Alexey A. Gritsenko, Anastasija Ili\\'c, Charles Blundell, Christos Kaplanis, Goker Erdogan, Ioana Bica, Jovana Mitrovi\\'c, Matko Bo\\v{s}njak, Matthias Bauer, Matthias Minderer, Razvan Pascanu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-18T10:28:45Z","title":"Improving fine-grained understanding in image-text pre-training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.09865","kind":"arxiv","version":1},"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:ede75be3e20cf5ca0a0ff84932459338aee95f69aad68738009a6c806782b515","target":"record","created_at":"2026-07-05T07:35:03Z","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":"b76d98e4e8d6d957f61600ebc992ca60932aebc5a662499ddf68dc1545aa908e","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-01-18T10:28:45Z","title_canon_sha256":"c44c46b5de2c29aa43cc9271d8f32e261eedf29451733038d5fda5bce1fed2ba"},"schema_version":"1.0","source":{"id":"2401.09865","kind":"arxiv","version":1}},"canonical_sha256":"a5817f05159c840f8fd156d135604ac3e7d9604589cd4eff6c419db22236b437","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a5817f05159c840f8fd156d135604ac3e7d9604589cd4eff6c419db22236b437","first_computed_at":"2026-07-05T07:35:03.492211Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:35:03.492211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8AYG7KdFAAsOSCoimf0aO4crrrEfNqFQljgJqXCp8FHph4B29pEAmgvnQ96QX9lmvYbSWxmLbIS031pJv3ReAg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:35:03.492645Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.09865","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ede75be3e20cf5ca0a0ff84932459338aee95f69aad68738009a6c806782b515","sha256:4372bae660d5498aa6d64cc8e613161ac074de508a212f9fb51f4c25757ba9c3"],"state_sha256":"276b5a5d4eebd662936c21a4a1625f210a9bc1debd1876b017b0d0801af4b92c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cRX1BqYvf+KaX7XvUPJVB9jf5JdqE/U98tfPjaOi1aP4UbcM2N1yVzoV97lpI3+j7/UsD6p72T9ikgF/J3LICQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T20:21:03.401275Z","bundle_sha256":"acfe33eae5a7ccc3356125e25afd3050064cd37cdc6976fcf6afb4ed76707057"}}