{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:7EC52GAZCBL3GD3KEERNFGM5TN","short_pith_number":"pith:7EC52GAZ","canonical_record":{"source":{"id":"2107.07075","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T02:12:20Z","cross_cats_sorted":[],"title_canon_sha256":"9691eb38b0ec9932800eb23797425801191307bf77d41449f94b658b2bf766c1","abstract_canon_sha256":"bc1b35bda4640d12cd3a17d9177f910f348283b600c61feb1c5fd7b905af06ac"},"schema_version":"1.0"},"canonical_sha256":"f905dd18191057b30f6a2122d2999d9b59f33e0a8aec2d43baae20725c8365ce","source":{"kind":"arxiv","id":"2107.07075","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.07075","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"arxiv_version","alias_value":"2107.07075v2","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07075","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_12","alias_value":"7EC52GAZCBL3","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_16","alias_value":"7EC52GAZCBL3GD3K","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_8","alias_value":"7EC52GAZ","created_at":"2026-07-05T05:55:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:7EC52GAZCBL3GD3KEERNFGM5TN","target":"record","payload":{"canonical_record":{"source":{"id":"2107.07075","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T02:12:20Z","cross_cats_sorted":[],"title_canon_sha256":"9691eb38b0ec9932800eb23797425801191307bf77d41449f94b658b2bf766c1","abstract_canon_sha256":"bc1b35bda4640d12cd3a17d9177f910f348283b600c61feb1c5fd7b905af06ac"},"schema_version":"1.0"},"canonical_sha256":"f905dd18191057b30f6a2122d2999d9b59f33e0a8aec2d43baae20725c8365ce","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:55:06.558411Z","signature_b64":"Y1+aHmi+4yYstdrLLfrbk2sgsA8b79fRJ7v1LLXhO9LpiCyC9lGIQUN2eeWctODLDXiLkx+l/+xs6m8obPUNDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f905dd18191057b30f6a2122d2999d9b59f33e0a8aec2d43baae20725c8365ce","last_reissued_at":"2026-07-05T05:55:06.557948Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:55:06.557948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2107.07075","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":"2026-07-05T05:55:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7ltbuZe9QGF6tIkiLNHYlLPCwEOXuqzSegOCzad56bNdr6RKON/N7x3kfX9BI3AGlh3T8WcdwmEDBw+makK9CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:14:30.068819Z"},"content_sha256":"4d93b4c66264cdfd80e404cf2267b42b608de3be2bf071f9c49ea87d191afab7","schema_version":"1.0","event_id":"sha256:4d93b4c66264cdfd80e404cf2267b42b608de3be2bf071f9c49ea87d191afab7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:7EC52GAZCBL3GD3KEERNFGM5TN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deep Learning on a Data Diet: Finding Important Examples Early in Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Gintare Karolina Dziugaite, Mansheej Paul, Surya Ganguli","submitted_at":"2021-07-15T02:12:20Z","abstract_excerpt":"Recent success in deep learning has partially been driven by training increasingly overparametrized networks on ever larger datasets. It is therefore natural to ask: how much of the data is superfluous, which examples are important for generalization, and how do we find them? In this work, we make the striking observation that, in standard vision datasets, simple scores averaged over several weight initializations can be used to identify important examples very early in training. We propose two such scores -- the Gradient Normed (GraNd) and the Error L2-Norm (EL2N) scores -- and demonstrate th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07075","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/2107.07075/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":"2026-07-05T05:55:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fwyIsmeV4x1duKIysa2mxBfgi+9qPlp5jltujEbcnwX5pL+qO+WXlf4BaUJh63IFBGqSsypNEqVthaDgEojzCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T10:14:30.069340Z"},"content_sha256":"389493db65a19e28569c0775915417cb2be6748aec292ac2749a8d4d5d609cc0","schema_version":"1.0","event_id":"sha256:389493db65a19e28569c0775915417cb2be6748aec292ac2749a8d4d5d609cc0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7EC52GAZCBL3GD3KEERNFGM5TN/bundle.json","state_url":"https://pith.science/pith/7EC52GAZCBL3GD3KEERNFGM5TN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7EC52GAZCBL3GD3KEERNFGM5TN/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":"2026-08-08T10:14:30Z","links":{"resolver":"https://pith.science/pith/7EC52GAZCBL3GD3KEERNFGM5TN","bundle":"https://pith.science/pith/7EC52GAZCBL3GD3KEERNFGM5TN/bundle.json","state":"https://pith.science/pith/7EC52GAZCBL3GD3KEERNFGM5TN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7EC52GAZCBL3GD3KEERNFGM5TN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:7EC52GAZCBL3GD3KEERNFGM5TN","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":"bc1b35bda4640d12cd3a17d9177f910f348283b600c61feb1c5fd7b905af06ac","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T02:12:20Z","title_canon_sha256":"9691eb38b0ec9932800eb23797425801191307bf77d41449f94b658b2bf766c1"},"schema_version":"1.0","source":{"id":"2107.07075","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.07075","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"arxiv_version","alias_value":"2107.07075v2","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07075","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_12","alias_value":"7EC52GAZCBL3","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_16","alias_value":"7EC52GAZCBL3GD3K","created_at":"2026-07-05T05:55:06Z"},{"alias_kind":"pith_short_8","alias_value":"7EC52GAZ","created_at":"2026-07-05T05:55:06Z"}],"graph_snapshots":[{"event_id":"sha256:389493db65a19e28569c0775915417cb2be6748aec292ac2749a8d4d5d609cc0","target":"graph","created_at":"2026-07-05T05:55:06Z","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/2107.07075/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent success in deep learning has partially been driven by training increasingly overparametrized networks on ever larger datasets. It is therefore natural to ask: how much of the data is superfluous, which examples are important for generalization, and how do we find them? In this work, we make the striking observation that, in standard vision datasets, simple scores averaged over several weight initializations can be used to identify important examples very early in training. We propose two such scores -- the Gradient Normed (GraNd) and the Error L2-Norm (EL2N) scores -- and demonstrate th","authors_text":"Gintare Karolina Dziugaite, Mansheej Paul, Surya Ganguli","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T02:12:20Z","title":"Deep Learning on a Data Diet: Finding Important Examples Early in Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07075","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":"sha256:4d93b4c66264cdfd80e404cf2267b42b608de3be2bf071f9c49ea87d191afab7","target":"record","created_at":"2026-07-05T05:55:06Z","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":"bc1b35bda4640d12cd3a17d9177f910f348283b600c61feb1c5fd7b905af06ac","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-15T02:12:20Z","title_canon_sha256":"9691eb38b0ec9932800eb23797425801191307bf77d41449f94b658b2bf766c1"},"schema_version":"1.0","source":{"id":"2107.07075","kind":"arxiv","version":2}},"canonical_sha256":"f905dd18191057b30f6a2122d2999d9b59f33e0a8aec2d43baae20725c8365ce","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f905dd18191057b30f6a2122d2999d9b59f33e0a8aec2d43baae20725c8365ce","first_computed_at":"2026-07-05T05:55:06.557948Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:55:06.557948Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y1+aHmi+4yYstdrLLfrbk2sgsA8b79fRJ7v1LLXhO9LpiCyC9lGIQUN2eeWctODLDXiLkx+l/+xs6m8obPUNDg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:55:06.558411Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.07075","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4d93b4c66264cdfd80e404cf2267b42b608de3be2bf071f9c49ea87d191afab7","sha256:389493db65a19e28569c0775915417cb2be6748aec292ac2749a8d4d5d609cc0"],"state_sha256":"a326e421ab8add381da5dc42fbf4bfd6643c31b695a8486b7ddecd4173081bfe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8iX+/sJUDwgil/L6aFiIzXYHf0ajLd/AKtGQ8FZ33b4sZ/R3GbkSL6iDHywBqjJ82LmC+a2jFShFjEYavjnWDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T10:14:30.074615Z","bundle_sha256":"bdd694dab2864560bcecab3913ec50bcaa1e59fe6f1ebca4f505535a6612932b"}}