{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NTDTDZKB32E7CKTMHVJOBSOO5H","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":"106398e5ca7b0b0616a6223b6256f748a1fdc32592babf76cab96f85dc7b85de","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-09-29T06:18:15Z","title_canon_sha256":"5188987404c1ded245af936a4239f4433f3fd7736dfb2e83f6eae422c18984fa"},"schema_version":"1.0","source":{"id":"2309.17002","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.17002","created_at":"2026-07-05T07:54:33Z"},{"alias_kind":"arxiv_version","alias_value":"2309.17002v2","created_at":"2026-07-05T07:54:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.17002","created_at":"2026-07-05T07:54:33Z"},{"alias_kind":"pith_short_12","alias_value":"NTDTDZKB32E7","created_at":"2026-07-05T07:54:33Z"},{"alias_kind":"pith_short_16","alias_value":"NTDTDZKB32E7CKTM","created_at":"2026-07-05T07:54:33Z"},{"alias_kind":"pith_short_8","alias_value":"NTDTDZKB","created_at":"2026-07-05T07:54:33Z"}],"graph_snapshots":[{"event_id":"sha256:e756751b069c904d08f49468f996062cd214f47dd5efdd881efc64a48faa0e9c","target":"graph","created_at":"2026-07-05T07:54:33Z","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/2309.17002/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Pre-training on large-scale datasets and then fine-tuning on downstream tasks have become a standard practice in deep learning. However, pre-training data often contain label noise that may adversely affect the generalization of the model. This paper aims to understand the nature of noise in pre-training datasets and to mitigate its impact on downstream tasks. More specifically, through extensive experiments of supervised pre-training models on synthetic noisy ImageNet-1K and YFCC15M datasets, we demonstrate that while slight noise in pre-training can benefit in-domain (ID) transfer performanc","authors_text":"Ankit Shah, Bhiksha Raj, Hao Chen, Hongxin Wei, Jindong Wang, Masashi Sugiyama, Ran Tao, Xing Xie","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-09-29T06:18:15Z","title":"Understanding and Mitigating the Label Noise in Pre-training on Downstream Tasks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.17002","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":"sha256:965a7467d0be13c2622dac522cddb0431119bb8cbc47d8c1a146b2b980d40bf7","target":"record","created_at":"2026-07-05T07:54:33Z","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":"106398e5ca7b0b0616a6223b6256f748a1fdc32592babf76cab96f85dc7b85de","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-09-29T06:18:15Z","title_canon_sha256":"5188987404c1ded245af936a4239f4433f3fd7736dfb2e83f6eae422c18984fa"},"schema_version":"1.0","source":{"id":"2309.17002","kind":"arxiv","version":2}},"canonical_sha256":"6cc731e541de89f12a6c3d52e0c9cee9fb68ae6e8106f4146c2f4b3d2d15bc51","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6cc731e541de89f12a6c3d52e0c9cee9fb68ae6e8106f4146c2f4b3d2d15bc51","first_computed_at":"2026-07-05T07:54:33.129621Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:54:33.129621Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ODtfC2mBwOcQvyHphFDPLju9nLHnmF4jrhI1hgKl+3z2yi6ge54yTMfgCW5z0mxj9ippiSAopaQHDLvHD42ZAg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:54:33.130113Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.17002","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:965a7467d0be13c2622dac522cddb0431119bb8cbc47d8c1a146b2b980d40bf7","sha256:e756751b069c904d08f49468f996062cd214f47dd5efdd881efc64a48faa0e9c"],"state_sha256":"041e4312258d60c63c9ca22546621c9b99db9b2e74fa92c6beb2e12fa418d93b"}