{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MVFQILJPE5THTC42PDSWUSOCBE","short_pith_number":"pith:MVFQILJP","schema_version":"1.0","canonical_sha256":"654b042d2f2766798b9a78e56a49c20930845cd2c51c30459afa7e0a2ea512ae","source":{"kind":"arxiv","id":"2605.31090","version":1},"attestation_state":"computed","paper":{"title":"On Revisiting Entropy for Identifying Mislabeled Images","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chunlei Li, Guanglu Dong, Jingliang Hu, Lichao Mou, Pengfei Li, Xiao Xiang Zhu, Yilei Shi, Zixuan Zheng","submitted_at":"2026-05-29T09:59:32Z","abstract_excerpt":"Mislabeled samples in training datasets severely degrade the performance of deep networks, as overparameterized models tend to memorize erroneous labels. We address this challenge by proposing a novel approach for mislabeled data detection that leverages training dynamics. Our method is grounded in the key observation that correctly labeled samples exhibit consistent entropy decrease during training, while mislabeled samples maintain relatively high entropy throughout the training process. Building on this insight, we introduce a signed entropy integral (SEI) statistic that captures both the m"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.31090","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-29T09:59:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5b27a915f1928973ffd660a119008d2eb515fcc1bd12ad2b65c4c8a7daa19d69","abstract_canon_sha256":"862db744a5b9a21f7389e6f7ee310c020206013a05505b3c67621daf649d8f19"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:57.358713Z","signature_b64":"D/VmwSGd46+IUsz10Q0UdMqfnRv3TwD55bvFl/Sk5hR+LTljiikICWM+q40qUN7gpICkkhcB+0X7DYQM0i4/BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"654b042d2f2766798b9a78e56a49c20930845cd2c51c30459afa7e0a2ea512ae","last_reissued_at":"2026-06-01T01:03:57.358057Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:57.358057Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"On Revisiting Entropy for Identifying Mislabeled Images","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chunlei Li, Guanglu Dong, Jingliang Hu, Lichao Mou, Pengfei Li, Xiao Xiang Zhu, Yilei Shi, Zixuan Zheng","submitted_at":"2026-05-29T09:59:32Z","abstract_excerpt":"Mislabeled samples in training datasets severely degrade the performance of deep networks, as overparameterized models tend to memorize erroneous labels. We address this challenge by proposing a novel approach for mislabeled data detection that leverages training dynamics. Our method is grounded in the key observation that correctly labeled samples exhibit consistent entropy decrease during training, while mislabeled samples maintain relatively high entropy throughout the training process. Building on this insight, we introduce a signed entropy integral (SEI) statistic that captures both the m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.31090","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/2605.31090/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2605.31090","created_at":"2026-06-01T01:03:57.358159+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.31090v1","created_at":"2026-06-01T01:03:57.358159+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.31090","created_at":"2026-06-01T01:03:57.358159+00:00"},{"alias_kind":"pith_short_12","alias_value":"MVFQILJPE5TH","created_at":"2026-06-01T01:03:57.358159+00:00"},{"alias_kind":"pith_short_16","alias_value":"MVFQILJPE5THTC42","created_at":"2026-06-01T01:03:57.358159+00:00"},{"alias_kind":"pith_short_8","alias_value":"MVFQILJP","created_at":"2026-06-01T01:03:57.358159+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE","json":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE.json","graph_json":"https://pith.science/api/pith-number/MVFQILJPE5THTC42PDSWUSOCBE/graph.json","events_json":"https://pith.science/api/pith-number/MVFQILJPE5THTC42PDSWUSOCBE/events.json","paper":"https://pith.science/paper/MVFQILJP"},"agent_actions":{"view_html":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE","download_json":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE.json","view_paper":"https://pith.science/paper/MVFQILJP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.31090&json=true","fetch_graph":"https://pith.science/api/pith-number/MVFQILJPE5THTC42PDSWUSOCBE/graph.json","fetch_events":"https://pith.science/api/pith-number/MVFQILJPE5THTC42PDSWUSOCBE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE/action/storage_attestation","attest_author":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE/action/author_attestation","sign_citation":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE/action/citation_signature","submit_replication":"https://pith.science/pith/MVFQILJPE5THTC42PDSWUSOCBE/action/replication_record"}},"created_at":"2026-06-01T01:03:57.358159+00:00","updated_at":"2026-06-01T01:03:57.358159+00:00"}