{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UL67CWOIPNA2ANC46AIJA5EKB2","short_pith_number":"pith:UL67CWOI","schema_version":"1.0","canonical_sha256":"a2fdf159c87b41a0345cf01090748a0e976c9a11babccb733ad2ed5e070e1e22","source":{"kind":"arxiv","id":"2505.03569","version":2},"attestation_state":"computed","paper":{"title":"Corner Cases: How Size and Position of Objects Challenge ImageNet-Trained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Margret Keuper, Mishal Fatima, Steffen Jung","submitted_at":"2025-05-06T14:27:01Z","abstract_excerpt":"Backgrounds in images play a major role in contributing to spurious correlations among different data points. Owing to aesthetic preferences of humans capturing the images, datasets can exhibit positional (location of the object within a given frame) and size (region-of-interest to image ratio) biases for different classes. In this paper, we show that these biases can impact how much a model relies on spurious features in the background to make its predictions. To better illustrate our findings, we propose a synthetic dataset derived from ImageNet-1k, Hard-Spurious-ImageNet, which contains ima"},"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":"2505.03569","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-06T14:27:01Z","cross_cats_sorted":[],"title_canon_sha256":"281bb191ac471e173edfcfe011f3c126ac834683fb0116079b7768fe203591df","abstract_canon_sha256":"ffc18cc4afd7e2466c9219bc09f81c3c271091fc23c9a5799a9b1a6ee4221c1b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:06:47.940945Z","signature_b64":"pqB2l3wEXo2Aa1sJetBj5V2zl+dWmknkwIzlylmjlxv0ZxEvKng/iEq3hpW0vql85X92cEieQV6L+yncmDy2CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a2fdf159c87b41a0345cf01090748a0e976c9a11babccb733ad2ed5e070e1e22","last_reissued_at":"2026-07-05T12:06:47.940495Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:06:47.940495Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Corner Cases: How Size and Position of Objects Challenge ImageNet-Trained Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Margret Keuper, Mishal Fatima, Steffen Jung","submitted_at":"2025-05-06T14:27:01Z","abstract_excerpt":"Backgrounds in images play a major role in contributing to spurious correlations among different data points. Owing to aesthetic preferences of humans capturing the images, datasets can exhibit positional (location of the object within a given frame) and size (region-of-interest to image ratio) biases for different classes. In this paper, we show that these biases can impact how much a model relies on spurious features in the background to make its predictions. To better illustrate our findings, we propose a synthetic dataset derived from ImageNet-1k, Hard-Spurious-ImageNet, which contains ima"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03569","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/2505.03569/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":"2505.03569","created_at":"2026-07-05T12:06:47.940565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03569v2","created_at":"2026-07-05T12:06:47.940565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03569","created_at":"2026-07-05T12:06:47.940565+00:00"},{"alias_kind":"pith_short_12","alias_value":"UL67CWOIPNA2","created_at":"2026-07-05T12:06:47.940565+00:00"},{"alias_kind":"pith_short_16","alias_value":"UL67CWOIPNA2ANC4","created_at":"2026-07-05T12:06:47.940565+00:00"},{"alias_kind":"pith_short_8","alias_value":"UL67CWOI","created_at":"2026-07-05T12:06:47.940565+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/UL67CWOIPNA2ANC46AIJA5EKB2","json":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2.json","graph_json":"https://pith.science/api/pith-number/UL67CWOIPNA2ANC46AIJA5EKB2/graph.json","events_json":"https://pith.science/api/pith-number/UL67CWOIPNA2ANC46AIJA5EKB2/events.json","paper":"https://pith.science/paper/UL67CWOI"},"agent_actions":{"view_html":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2","download_json":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2.json","view_paper":"https://pith.science/paper/UL67CWOI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03569&json=true","fetch_graph":"https://pith.science/api/pith-number/UL67CWOIPNA2ANC46AIJA5EKB2/graph.json","fetch_events":"https://pith.science/api/pith-number/UL67CWOIPNA2ANC46AIJA5EKB2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2/action/storage_attestation","attest_author":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2/action/author_attestation","sign_citation":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2/action/citation_signature","submit_replication":"https://pith.science/pith/UL67CWOIPNA2ANC46AIJA5EKB2/action/replication_record"}},"created_at":"2026-07-05T12:06:47.940565+00:00","updated_at":"2026-07-05T12:06:47.940565+00:00"}