{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:64ONE652SWLDW2EONR5IZBM24O","short_pith_number":"pith:64ONE652","schema_version":"1.0","canonical_sha256":"f71cd27bba95963b688e6c7a8c859ae3899151441329aeed0991d605d36b4336","source":{"kind":"arxiv","id":"2606.11319","version":1},"attestation_state":"computed","paper":{"title":"Learning from almost nothing: How neural networks survive heavy input corruption","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Asadullah Bhuiyan, Hyejin Kim, Justin Tahmassebpur, Omri Lesser","submitted_at":"2026-06-09T18:02:09Z","abstract_excerpt":"Learning from imperfect data is a central theme in machine learning, connecting practical questions of robustness to fundamental questions of learnability. Here we examine attribute noise: learning from corrupted inputs while keeping the labels intact, a setting that has received considerably less analytical attention than its label-noise counterpart. We consider two types of corruption models: additive noise and replacement noise. Through experiments with multi-layer perceptrons (MLPs) on corrupted classification datasets, we find that neural networks remain robust, maintaining well-above-cha"},"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":"2606.11319","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-09T18:02:09Z","cross_cats_sorted":["cond-mat.dis-nn"],"title_canon_sha256":"972eae5bf24c8c6da913cd2f81eb36c67e6fc4bf8c020014c2309de6b0e2e54d","abstract_canon_sha256":"b77a5ecfffea26e1f445d027b249f64cc097351c54a88a988229471bd40a86e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T00:08:19.180438Z","signature_b64":"iLgfDzetIG4hhYToUEamAuiiDIX/oUS7aGSYnj8KNjkSy61HGEk4MzO9KvsFvDkjt/ECJoip4ZohjhfKwXlPDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f71cd27bba95963b688e6c7a8c859ae3899151441329aeed0991d605d36b4336","last_reissued_at":"2026-06-11T00:08:19.179736Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T00:08:19.179736Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning from almost nothing: How neural networks survive heavy input corruption","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.dis-nn"],"primary_cat":"cs.LG","authors_text":"Asadullah Bhuiyan, Hyejin Kim, Justin Tahmassebpur, Omri Lesser","submitted_at":"2026-06-09T18:02:09Z","abstract_excerpt":"Learning from imperfect data is a central theme in machine learning, connecting practical questions of robustness to fundamental questions of learnability. Here we examine attribute noise: learning from corrupted inputs while keeping the labels intact, a setting that has received considerably less analytical attention than its label-noise counterpart. We consider two types of corruption models: additive noise and replacement noise. Through experiments with multi-layer perceptrons (MLPs) on corrupted classification datasets, we find that neural networks remain robust, maintaining well-above-cha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.11319","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/2606.11319/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":"2606.11319","created_at":"2026-06-11T00:08:19.179840+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.11319v1","created_at":"2026-06-11T00:08:19.179840+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.11319","created_at":"2026-06-11T00:08:19.179840+00:00"},{"alias_kind":"pith_short_12","alias_value":"64ONE652SWLD","created_at":"2026-06-11T00:08:19.179840+00:00"},{"alias_kind":"pith_short_16","alias_value":"64ONE652SWLDW2EO","created_at":"2026-06-11T00:08:19.179840+00:00"},{"alias_kind":"pith_short_8","alias_value":"64ONE652","created_at":"2026-06-11T00:08:19.179840+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/64ONE652SWLDW2EONR5IZBM24O","json":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O.json","graph_json":"https://pith.science/api/pith-number/64ONE652SWLDW2EONR5IZBM24O/graph.json","events_json":"https://pith.science/api/pith-number/64ONE652SWLDW2EONR5IZBM24O/events.json","paper":"https://pith.science/paper/64ONE652"},"agent_actions":{"view_html":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O","download_json":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O.json","view_paper":"https://pith.science/paper/64ONE652","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.11319&json=true","fetch_graph":"https://pith.science/api/pith-number/64ONE652SWLDW2EONR5IZBM24O/graph.json","fetch_events":"https://pith.science/api/pith-number/64ONE652SWLDW2EONR5IZBM24O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O/action/storage_attestation","attest_author":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O/action/author_attestation","sign_citation":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O/action/citation_signature","submit_replication":"https://pith.science/pith/64ONE652SWLDW2EONR5IZBM24O/action/replication_record"}},"created_at":"2026-06-11T00:08:19.179840+00:00","updated_at":"2026-06-11T00:08:19.179840+00:00"}