{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:AIVMTH6AEDP2GFQPMBKVF42IWE","short_pith_number":"pith:AIVMTH6A","schema_version":"1.0","canonical_sha256":"022ac99fc020dfa3160f605552f348b1101067c4ef4bc81c2c1136e60065e876","source":{"kind":"arxiv","id":"2210.13007","version":2},"attestation_state":"computed","paper":{"title":"Iterative Patch Selection for High-Resolution Image Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Aravindh Mahendran, Benjamin Bergner, Christoph Lippert","submitted_at":"2022-10-24T07:55:57Z","abstract_excerpt":"High-resolution images are prevalent in various applications, such as autonomous driving and computer-aided diagnosis. However, training neural networks on such images is computationally challenging and easily leads to out-of-memory errors even on modern GPUs. We propose a simple method, Iterative Patch Selection (IPS), which decouples the memory usage from the input size and thus enables the processing of arbitrarily large images under tight hardware constraints. IPS achieves this by selecting only the most salient patches, which are then aggregated into a global representation for image reco"},"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":"2210.13007","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-10-24T07:55:57Z","cross_cats_sorted":["cs.LG","eess.IV"],"title_canon_sha256":"ccc472322ef73656b1b73bc4f92f1353ddfe2016ce079834ff2bf3b85fd3fafd","abstract_canon_sha256":"6a3f07f8af7d3088bad671fa15b46e275325747c1cb75292bc622204af55ae4a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:48:41.009858Z","signature_b64":"V7mvivh8Gl6em046+jXvX259c2KHi2EqZyId9YjhE7Dbj2Vj2WH3XyLPvKt2KmaGmB4zZNuUtzN0WgsyW79iBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"022ac99fc020dfa3160f605552f348b1101067c4ef4bc81c2c1136e60065e876","last_reissued_at":"2026-07-05T05:48:41.009371Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:48:41.009371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Iterative Patch Selection for High-Resolution Image Recognition","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Aravindh Mahendran, Benjamin Bergner, Christoph Lippert","submitted_at":"2022-10-24T07:55:57Z","abstract_excerpt":"High-resolution images are prevalent in various applications, such as autonomous driving and computer-aided diagnosis. However, training neural networks on such images is computationally challenging and easily leads to out-of-memory errors even on modern GPUs. We propose a simple method, Iterative Patch Selection (IPS), which decouples the memory usage from the input size and thus enables the processing of arbitrarily large images under tight hardware constraints. IPS achieves this by selecting only the most salient patches, which are then aggregated into a global representation for image reco"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.13007","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/2210.13007/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":"2210.13007","created_at":"2026-07-05T05:48:41.009435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.13007v2","created_at":"2026-07-05T05:48:41.009435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.13007","created_at":"2026-07-05T05:48:41.009435+00:00"},{"alias_kind":"pith_short_12","alias_value":"AIVMTH6AEDP2","created_at":"2026-07-05T05:48:41.009435+00:00"},{"alias_kind":"pith_short_16","alias_value":"AIVMTH6AEDP2GFQP","created_at":"2026-07-05T05:48:41.009435+00:00"},{"alias_kind":"pith_short_8","alias_value":"AIVMTH6A","created_at":"2026-07-05T05:48:41.009435+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.11237","citing_title":"On the Generalizability of Iterative Patch Selection for Memory-Efficient High-Resolution Image Classification","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE","json":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE.json","graph_json":"https://pith.science/api/pith-number/AIVMTH6AEDP2GFQPMBKVF42IWE/graph.json","events_json":"https://pith.science/api/pith-number/AIVMTH6AEDP2GFQPMBKVF42IWE/events.json","paper":"https://pith.science/paper/AIVMTH6A"},"agent_actions":{"view_html":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE","download_json":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE.json","view_paper":"https://pith.science/paper/AIVMTH6A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.13007&json=true","fetch_graph":"https://pith.science/api/pith-number/AIVMTH6AEDP2GFQPMBKVF42IWE/graph.json","fetch_events":"https://pith.science/api/pith-number/AIVMTH6AEDP2GFQPMBKVF42IWE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE/action/storage_attestation","attest_author":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE/action/author_attestation","sign_citation":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE/action/citation_signature","submit_replication":"https://pith.science/pith/AIVMTH6AEDP2GFQPMBKVF42IWE/action/replication_record"}},"created_at":"2026-07-05T05:48:41.009435+00:00","updated_at":"2026-07-05T05:48:41.009435+00:00"}