{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:GJQS5UKCCJT7UD5L43TYQD5RRG","short_pith_number":"pith:GJQS5UKC","schema_version":"1.0","canonical_sha256":"32612ed1421267fa0fabe6e7880fb18996a6cda9e20a2fb18a78b2ef773a0173","source":{"kind":"arxiv","id":"2104.00795","version":1},"attestation_state":"computed","paper":{"title":"No Cost Likelihood Manipulation at Test Time for Making Better Mistakes in Deep Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ameya Prabhu, Puneet K. Dokania, Shyamgopal Karthik, Vineet Gandhi","submitted_at":"2021-04-01T22:40:25Z","abstract_excerpt":"There has been increasing interest in building deep hierarchy-aware classifiers that aim to quantify and reduce the severity of mistakes, and not just reduce the number of errors. The idea is to exploit the label hierarchy (e.g., the WordNet ontology) and consider graph distances as a proxy for mistake severity. Surprisingly, on examining mistake-severity distributions of the top-1 prediction, we find that current state-of-the-art hierarchy-aware deep classifiers do not always show practical improvement over the standard cross-entropy baseline in making better mistakes. The reason for the redu"},"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":"2104.00795","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2021-04-01T22:40:25Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"2fb80bdf3ee74af9abb47217e74593eed4bb67d0b236a4d6e6e69f1f413f7b07","abstract_canon_sha256":"2bd566aaf8d8379f0cc8107fc4206a8b99b84c2bf297570671c9a68407ee6845"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:28:35.981829Z","signature_b64":"d3prvgq4jfA7uMSofAr3fn2Vr47IkeNM5rM9g1tzJv0PYICSPO74uu9k7AFDpM5/E+S1QEpPymaqgn8KtNWqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32612ed1421267fa0fabe6e7880fb18996a6cda9e20a2fb18a78b2ef773a0173","last_reissued_at":"2026-07-05T02:28:35.981426Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:28:35.981426Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"No Cost Likelihood Manipulation at Test Time for Making Better Mistakes in Deep Networks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ameya Prabhu, Puneet K. Dokania, Shyamgopal Karthik, Vineet Gandhi","submitted_at":"2021-04-01T22:40:25Z","abstract_excerpt":"There has been increasing interest in building deep hierarchy-aware classifiers that aim to quantify and reduce the severity of mistakes, and not just reduce the number of errors. The idea is to exploit the label hierarchy (e.g., the WordNet ontology) and consider graph distances as a proxy for mistake severity. Surprisingly, on examining mistake-severity distributions of the top-1 prediction, we find that current state-of-the-art hierarchy-aware deep classifiers do not always show practical improvement over the standard cross-entropy baseline in making better mistakes. The reason for the redu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.00795","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/2104.00795/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":"2104.00795","created_at":"2026-07-05T02:28:35.981482+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.00795v1","created_at":"2026-07-05T02:28:35.981482+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.00795","created_at":"2026-07-05T02:28:35.981482+00:00"},{"alias_kind":"pith_short_12","alias_value":"GJQS5UKCCJT7","created_at":"2026-07-05T02:28:35.981482+00:00"},{"alias_kind":"pith_short_16","alias_value":"GJQS5UKCCJT7UD5L","created_at":"2026-07-05T02:28:35.981482+00:00"},{"alias_kind":"pith_short_8","alias_value":"GJQS5UKC","created_at":"2026-07-05T02:28:35.981482+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/GJQS5UKCCJT7UD5L43TYQD5RRG","json":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG.json","graph_json":"https://pith.science/api/pith-number/GJQS5UKCCJT7UD5L43TYQD5RRG/graph.json","events_json":"https://pith.science/api/pith-number/GJQS5UKCCJT7UD5L43TYQD5RRG/events.json","paper":"https://pith.science/paper/GJQS5UKC"},"agent_actions":{"view_html":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG","download_json":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG.json","view_paper":"https://pith.science/paper/GJQS5UKC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.00795&json=true","fetch_graph":"https://pith.science/api/pith-number/GJQS5UKCCJT7UD5L43TYQD5RRG/graph.json","fetch_events":"https://pith.science/api/pith-number/GJQS5UKCCJT7UD5L43TYQD5RRG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG/action/storage_attestation","attest_author":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG/action/author_attestation","sign_citation":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG/action/citation_signature","submit_replication":"https://pith.science/pith/GJQS5UKCCJT7UD5L43TYQD5RRG/action/replication_record"}},"created_at":"2026-07-05T02:28:35.981482+00:00","updated_at":"2026-07-05T02:28:35.981482+00:00"}