{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:B753YM4NZEJ7IPDCT6VK33246D","short_pith_number":"pith:B753YM4N","canonical_record":{"source":{"id":"2303.17127","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T03:22:52Z","cross_cats_sorted":[],"title_canon_sha256":"75474042d6f7b7a34ce29b17f3f4894b02cc2b65bf9280a9ca5fbae653579c3e","abstract_canon_sha256":"b0774ef3c6bf1cf78a8fbd4dffdf96b25403e27af58c26a75704df25e22e6284"},"schema_version":"1.0"},"canonical_sha256":"0ffbbc338dc913f43c629faaadef5cf0c1435da959aa02b2711306acd4b82c7b","source":{"kind":"arxiv","id":"2303.17127","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.17127","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"arxiv_version","alias_value":"2303.17127v1","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.17127","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_12","alias_value":"B753YM4NZEJ7","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_16","alias_value":"B753YM4NZEJ7IPDC","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_8","alias_value":"B753YM4N","created_at":"2026-07-05T05:56:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:B753YM4NZEJ7IPDCT6VK33246D","target":"record","payload":{"canonical_record":{"source":{"id":"2303.17127","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T03:22:52Z","cross_cats_sorted":[],"title_canon_sha256":"75474042d6f7b7a34ce29b17f3f4894b02cc2b65bf9280a9ca5fbae653579c3e","abstract_canon_sha256":"b0774ef3c6bf1cf78a8fbd4dffdf96b25403e27af58c26a75704df25e22e6284"},"schema_version":"1.0"},"canonical_sha256":"0ffbbc338dc913f43c629faaadef5cf0c1435da959aa02b2711306acd4b82c7b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:56:19.888570Z","signature_b64":"Cl+82oiuIvyZYRkITJ5MvzbU9KsoILb73Jo/ov50p5hsEMkN/4YrrxL2lcBecHLTZIidX7HB0HPaqgYwDUtnDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0ffbbc338dc913f43c629faaadef5cf0c1435da959aa02b2711306acd4b82c7b","last_reissued_at":"2026-07-05T05:56:19.888023Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:56:19.888023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2303.17127","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":"2026-07-05T05:56:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uzOfsdL2oOtfT3fJlkfyjHi+hwxpS1mNuNdZYzY/Cy18O4uLm1e2RMD5iFwYoTnohBy1fuRi460Rk+Ehx8f4Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:38:55.844700Z"},"content_sha256":"52e55f9dde92f4d6e5add593b47b0030d471679ccc55f55ec2d6c300841739a4","schema_version":"1.0","event_id":"sha256:52e55f9dde92f4d6e5add593b47b0030d471679ccc55f55ec2d6c300841739a4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:B753YM4NZEJ7IPDCT6VK33246D","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adaptive Cross Batch Normalization for Metric Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Anton van den Hengel, Matt Ma, Stephen Gould, Thalaiyasingam Ajanthan","submitted_at":"2023-03-30T03:22:52Z","abstract_excerpt":"Metric learning is a fundamental problem in computer vision whereby a model is trained to learn a semantically useful embedding space via ranking losses. Traditionally, the effectiveness of a ranking loss depends on the minibatch size, and is, therefore, inherently limited by the memory constraints of the underlying hardware. While simply accumulating the embeddings across minibatches has proved useful (Wang et al. [2020]), we show that it is equally important to ensure that the accumulated embeddings are up to date. In particular, it is necessary to circumvent the representational drift betwe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.17127","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/2303.17127/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":"2026-07-05T05:56:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8xGWxCIzlBefcFuqyPD6ED0g4kSxzjgDsL+d8zUgqhhvKEUevB7G7riyHiaBYbzCjqyZys80Mt3d5sQjsd+FCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:38:55.845093Z"},"content_sha256":"f8353e7284593f5b2e551f8a1a2ecdaffe09ef76f107b4184e336393431ec76c","schema_version":"1.0","event_id":"sha256:f8353e7284593f5b2e551f8a1a2ecdaffe09ef76f107b4184e336393431ec76c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B753YM4NZEJ7IPDCT6VK33246D/bundle.json","state_url":"https://pith.science/pith/B753YM4NZEJ7IPDCT6VK33246D/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B753YM4NZEJ7IPDCT6VK33246D/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":"2026-07-07T06:38:55Z","links":{"resolver":"https://pith.science/pith/B753YM4NZEJ7IPDCT6VK33246D","bundle":"https://pith.science/pith/B753YM4NZEJ7IPDCT6VK33246D/bundle.json","state":"https://pith.science/pith/B753YM4NZEJ7IPDCT6VK33246D/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B753YM4NZEJ7IPDCT6VK33246D/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:B753YM4NZEJ7IPDCT6VK33246D","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":"b0774ef3c6bf1cf78a8fbd4dffdf96b25403e27af58c26a75704df25e22e6284","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T03:22:52Z","title_canon_sha256":"75474042d6f7b7a34ce29b17f3f4894b02cc2b65bf9280a9ca5fbae653579c3e"},"schema_version":"1.0","source":{"id":"2303.17127","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.17127","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"arxiv_version","alias_value":"2303.17127v1","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.17127","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_12","alias_value":"B753YM4NZEJ7","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_16","alias_value":"B753YM4NZEJ7IPDC","created_at":"2026-07-05T05:56:19Z"},{"alias_kind":"pith_short_8","alias_value":"B753YM4N","created_at":"2026-07-05T05:56:19Z"}],"graph_snapshots":[{"event_id":"sha256:f8353e7284593f5b2e551f8a1a2ecdaffe09ef76f107b4184e336393431ec76c","target":"graph","created_at":"2026-07-05T05:56:19Z","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/2303.17127/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Metric learning is a fundamental problem in computer vision whereby a model is trained to learn a semantically useful embedding space via ranking losses. Traditionally, the effectiveness of a ranking loss depends on the minibatch size, and is, therefore, inherently limited by the memory constraints of the underlying hardware. While simply accumulating the embeddings across minibatches has proved useful (Wang et al. [2020]), we show that it is equally important to ensure that the accumulated embeddings are up to date. In particular, it is necessary to circumvent the representational drift betwe","authors_text":"Anton van den Hengel, Matt Ma, Stephen Gould, Thalaiyasingam Ajanthan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T03:22:52Z","title":"Adaptive Cross Batch Normalization for Metric Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2303.17127","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":"sha256:52e55f9dde92f4d6e5add593b47b0030d471679ccc55f55ec2d6c300841739a4","target":"record","created_at":"2026-07-05T05:56:19Z","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":"b0774ef3c6bf1cf78a8fbd4dffdf96b25403e27af58c26a75704df25e22e6284","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-03-30T03:22:52Z","title_canon_sha256":"75474042d6f7b7a34ce29b17f3f4894b02cc2b65bf9280a9ca5fbae653579c3e"},"schema_version":"1.0","source":{"id":"2303.17127","kind":"arxiv","version":1}},"canonical_sha256":"0ffbbc338dc913f43c629faaadef5cf0c1435da959aa02b2711306acd4b82c7b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0ffbbc338dc913f43c629faaadef5cf0c1435da959aa02b2711306acd4b82c7b","first_computed_at":"2026-07-05T05:56:19.888023Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:56:19.888023Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Cl+82oiuIvyZYRkITJ5MvzbU9KsoILb73Jo/ov50p5hsEMkN/4YrrxL2lcBecHLTZIidX7HB0HPaqgYwDUtnDA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:56:19.888570Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.17127","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52e55f9dde92f4d6e5add593b47b0030d471679ccc55f55ec2d6c300841739a4","sha256:f8353e7284593f5b2e551f8a1a2ecdaffe09ef76f107b4184e336393431ec76c"],"state_sha256":"3b3cac803909c9c8619a9838c5ed70b3ff83c2c0203291c5260dccac2eba1e55"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Sd1bg/0/g/4fAb4qVAxkbjrtWgepdny83BieDysADMT3uWp9srlEWJdnh6BRAgNQXgQOrKhc+pqs1hrJ+ZRYDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:38:55.847443Z","bundle_sha256":"6bb1c94bc6c605b8485e65e0113ae5bdb55beeae5502972e035be7566d9849a3"}}