{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:SJUSKOWTTEN7DIS2MQ3Y7QFOCD","short_pith_number":"pith:SJUSKOWT","schema_version":"1.0","canonical_sha256":"9269253ad3991bf1a25a64378fc0ae10e6c0e0360592ce323d2e3a4c30f484a1","source":{"kind":"arxiv","id":"1909.04868","version":8},"attestation_state":"computed","paper":{"title":"Is Heuristic Sampling Necessary in Training Deep Object Detectors?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Liu, Enhong Chen, Joya Chen, Shiwei Wu, Tong Xu, Yifei Cheng","submitted_at":"2019-09-11T06:29:58Z","abstract_excerpt":"To train accurate deep object detectors under the extreme foreground-background imbalance, heuristic sampling methods are always necessary, which either re-sample a subset of all training samples (hard sampling methods, \\eg biased sampling, OHEM), or use all training samples but re-weight them discriminatively (soft sampling methods, \\eg Focal Loss, GHM). In this paper, we challenge the necessity of such hard/soft sampling methods for training accurate deep object detectors. While previous studies have shown that training detectors without heuristic sampling methods would significantly degrade"},"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":"1909.04868","kind":"arxiv","version":8},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-09-11T06:29:58Z","cross_cats_sorted":[],"title_canon_sha256":"f79a7192cd61bc9daa5b147df6ff438dca6f3721fca40e80ea6e9665ba0a9de9","abstract_canon_sha256":"d09e53cf8d501a789fb1ed5ca3b807c0383db72c2270af0249b5085d9d0635ba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:04:58.461382Z","signature_b64":"M8FTaTyYgqZyGVC/As61F9RD1g9odJW1x5wVDJ5SQhtx5hGAurCa1F65Z0eTCLrD/xKMTwqfkAAIRQzeVqWyDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9269253ad3991bf1a25a64378fc0ae10e6c0e0360592ce323d2e3a4c30f484a1","last_reissued_at":"2026-07-05T03:04:58.460787Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:04:58.460787Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Heuristic Sampling Necessary in Training Deep Object Detectors?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dong Liu, Enhong Chen, Joya Chen, Shiwei Wu, Tong Xu, Yifei Cheng","submitted_at":"2019-09-11T06:29:58Z","abstract_excerpt":"To train accurate deep object detectors under the extreme foreground-background imbalance, heuristic sampling methods are always necessary, which either re-sample a subset of all training samples (hard sampling methods, \\eg biased sampling, OHEM), or use all training samples but re-weight them discriminatively (soft sampling methods, \\eg Focal Loss, GHM). In this paper, we challenge the necessity of such hard/soft sampling methods for training accurate deep object detectors. While previous studies have shown that training detectors without heuristic sampling methods would significantly degrade"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1909.04868","kind":"arxiv","version":8},"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/1909.04868/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":"1909.04868","created_at":"2026-07-05T03:04:58.460861+00:00"},{"alias_kind":"arxiv_version","alias_value":"1909.04868v8","created_at":"2026-07-05T03:04:58.460861+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1909.04868","created_at":"2026-07-05T03:04:58.460861+00:00"},{"alias_kind":"pith_short_12","alias_value":"SJUSKOWTTEN7","created_at":"2026-07-05T03:04:58.460861+00:00"},{"alias_kind":"pith_short_16","alias_value":"SJUSKOWTTEN7DIS2","created_at":"2026-07-05T03:04:58.460861+00:00"},{"alias_kind":"pith_short_8","alias_value":"SJUSKOWT","created_at":"2026-07-05T03:04:58.460861+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/SJUSKOWTTEN7DIS2MQ3Y7QFOCD","json":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD.json","graph_json":"https://pith.science/api/pith-number/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/graph.json","events_json":"https://pith.science/api/pith-number/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/events.json","paper":"https://pith.science/paper/SJUSKOWT"},"agent_actions":{"view_html":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD","download_json":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD.json","view_paper":"https://pith.science/paper/SJUSKOWT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1909.04868&json=true","fetch_graph":"https://pith.science/api/pith-number/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/graph.json","fetch_events":"https://pith.science/api/pith-number/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/action/storage_attestation","attest_author":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/action/author_attestation","sign_citation":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/action/citation_signature","submit_replication":"https://pith.science/pith/SJUSKOWTTEN7DIS2MQ3Y7QFOCD/action/replication_record"}},"created_at":"2026-07-05T03:04:58.460861+00:00","updated_at":"2026-07-05T03:04:58.460861+00:00"}