{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DJCPEPJ6SNZBDVWLTRX6HH3UNH","short_pith_number":"pith:DJCPEPJ6","canonical_record":{"source":{"id":"2405.21070","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T17:57:24Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"e4829f476edde733a2d7f989e981211122ab0d9ec03544e2f0c83675ce38caae","abstract_canon_sha256":"0f90962eb59a9fd812c1b92216efc67348c416d2b5025c124fa6835c77cb8003"},"schema_version":"1.0"},"canonical_sha256":"1a44f23d3e937211d6cb9c6fe39f7469f23a495526904ea3e9ef14d0400180e8","source":{"kind":"arxiv","id":"2405.21070","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.21070","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"arxiv_version","alias_value":"2405.21070v3","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.21070","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_12","alias_value":"DJCPEPJ6SNZB","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_16","alias_value":"DJCPEPJ6SNZBDVWL","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_8","alias_value":"DJCPEPJ6","created_at":"2026-07-05T09:27:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DJCPEPJ6SNZBDVWLTRX6HH3UNH","target":"record","payload":{"canonical_record":{"source":{"id":"2405.21070","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T17:57:24Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"e4829f476edde733a2d7f989e981211122ab0d9ec03544e2f0c83675ce38caae","abstract_canon_sha256":"0f90962eb59a9fd812c1b92216efc67348c416d2b5025c124fa6835c77cb8003"},"schema_version":"1.0"},"canonical_sha256":"1a44f23d3e937211d6cb9c6fe39f7469f23a495526904ea3e9ef14d0400180e8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:01.455340Z","signature_b64":"vl7gglc+XS/G25W057soQTApiQCACl5BoJVQgP87I9JDlVb6kGqX3JW2jRoj1NEbDbUqshyj/4DXiB4dkd71Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1a44f23d3e937211d6cb9c6fe39f7469f23a495526904ea3e9ef14d0400180e8","last_reissued_at":"2026-07-05T09:27:01.454910Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:01.454910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.21070","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:27:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CYXsMV3oSfzx8Aouzxf1kkKTRzARYZJxhJHA9KmLdYARj4HmvgrsEA6wPZdjuA4V+rhiU8u3xVi6HZ8iwYDVDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:17:18.180173Z"},"content_sha256":"0dc374314f5fc014ade08892c63a9edd60678adb6f8e2601cc937e451ecb9ea3","schema_version":"1.0","event_id":"sha256:0dc374314f5fc014ade08892c63a9edd60678adb6f8e2601cc937e451ecb9ea3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DJCPEPJ6SNZBDVWLTRX6HH3UNH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Makes CLIP More Robust to Long-Tailed Pre-Training Data? A Controlled Study for Transferable Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Bingchen Zhao, Jiangmiao Pang, Xiaojuan Qi, Xin Wen, Yilun Chen","submitted_at":"2024-05-31T17:57:24Z","abstract_excerpt":"Severe data imbalance naturally exists among web-scale vision-language datasets. Despite this, we find CLIP pre-trained thereupon exhibits notable robustness to the data imbalance compared to supervised learning, and demonstrates significant effectiveness in learning generalizable representations. With an aim to investigate the reasons behind this finding, we conduct controlled experiments to study various underlying factors, and reveal that CLIP's pretext task forms a dynamic classification problem wherein only a subset of classes is present in training. This isolates the bias from dominant c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.21070","kind":"arxiv","version":3},"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/2405.21070/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":"2026-07-05T09:27:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7Ar8GDzNOAXWMoXNzGzw1tECNL6MOk8JlN3qTGjmeaZeBwehR1++/ECJi4U9EWCrfmVYzUk7VPMGz2/OUTuLDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:17:18.180687Z"},"content_sha256":"27b9ee9edf22225ebd221b52141b37e94ed6458c369414c5a46a43a2f1162aaa","schema_version":"1.0","event_id":"sha256:27b9ee9edf22225ebd221b52141b37e94ed6458c369414c5a46a43a2f1162aaa"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/bundle.json","state_url":"https://pith.science/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/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":"2026-08-03T16:17:18Z","links":{"resolver":"https://pith.science/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH","bundle":"https://pith.science/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/bundle.json","state":"https://pith.science/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DJCPEPJ6SNZBDVWLTRX6HH3UNH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DJCPEPJ6SNZBDVWLTRX6HH3UNH","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":"0f90962eb59a9fd812c1b92216efc67348c416d2b5025c124fa6835c77cb8003","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T17:57:24Z","title_canon_sha256":"e4829f476edde733a2d7f989e981211122ab0d9ec03544e2f0c83675ce38caae"},"schema_version":"1.0","source":{"id":"2405.21070","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.21070","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"arxiv_version","alias_value":"2405.21070v3","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.21070","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_12","alias_value":"DJCPEPJ6SNZB","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_16","alias_value":"DJCPEPJ6SNZBDVWL","created_at":"2026-07-05T09:27:01Z"},{"alias_kind":"pith_short_8","alias_value":"DJCPEPJ6","created_at":"2026-07-05T09:27:01Z"}],"graph_snapshots":[{"event_id":"sha256:27b9ee9edf22225ebd221b52141b37e94ed6458c369414c5a46a43a2f1162aaa","target":"graph","created_at":"2026-07-05T09:27:01Z","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/2405.21070/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Severe data imbalance naturally exists among web-scale vision-language datasets. Despite this, we find CLIP pre-trained thereupon exhibits notable robustness to the data imbalance compared to supervised learning, and demonstrates significant effectiveness in learning generalizable representations. With an aim to investigate the reasons behind this finding, we conduct controlled experiments to study various underlying factors, and reveal that CLIP's pretext task forms a dynamic classification problem wherein only a subset of classes is present in training. This isolates the bias from dominant c","authors_text":"Bingchen Zhao, Jiangmiao Pang, Xiaojuan Qi, Xin Wen, Yilun Chen","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T17:57:24Z","title":"What Makes CLIP More Robust to Long-Tailed Pre-Training Data? A Controlled Study for Transferable Insights"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.21070","kind":"arxiv","version":3},"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":"sha256:0dc374314f5fc014ade08892c63a9edd60678adb6f8e2601cc937e451ecb9ea3","target":"record","created_at":"2026-07-05T09:27:01Z","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":"0f90962eb59a9fd812c1b92216efc67348c416d2b5025c124fa6835c77cb8003","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-05-31T17:57:24Z","title_canon_sha256":"e4829f476edde733a2d7f989e981211122ab0d9ec03544e2f0c83675ce38caae"},"schema_version":"1.0","source":{"id":"2405.21070","kind":"arxiv","version":3}},"canonical_sha256":"1a44f23d3e937211d6cb9c6fe39f7469f23a495526904ea3e9ef14d0400180e8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1a44f23d3e937211d6cb9c6fe39f7469f23a495526904ea3e9ef14d0400180e8","first_computed_at":"2026-07-05T09:27:01.454910Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:01.454910Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vl7gglc+XS/G25W057soQTApiQCACl5BoJVQgP87I9JDlVb6kGqX3JW2jRoj1NEbDbUqshyj/4DXiB4dkd71Bw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:01.455340Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.21070","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0dc374314f5fc014ade08892c63a9edd60678adb6f8e2601cc937e451ecb9ea3","sha256:27b9ee9edf22225ebd221b52141b37e94ed6458c369414c5a46a43a2f1162aaa"],"state_sha256":"4ed38e28da1db964895fee519ac50901d3fb6d7d38aa83afd47432d436543274"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6ppJwhykn1l9QkeCHZTpL49j8BqnU7d2QAubXCCfaNhWyS7RqY5IDH+3EsTeFOeyNaF4bIrRo1/T/OUP7b+gAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:17:18.185112Z","bundle_sha256":"bcd932aa592323d8448a344c79829cba56ad16c7e0da02441d4176911e4604b0"}}