{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:T3L2GWZ34R4IO3JQ4KNKN3GE7U","short_pith_number":"pith:T3L2GWZ3","schema_version":"1.0","canonical_sha256":"9ed7a35b3be478876d30e29aa6ecc4fd3db9ad08162cf777bba49011442ccb39","source":{"kind":"arxiv","id":"2608.03681","version":1},"attestation_state":"computed","paper":{"title":"Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jingyuan Zhang, Ke Xu, Qixiang Ma, Sihang Zhou, Xiaoyun Wang, Yanning Hou","submitted_at":"2026-08-04T13:53:36Z","abstract_excerpt":"Zero-shot visual anomaly detection has achieved remarkable progress, with recent vision-only approaches further improving performance while simplifying the inference pipeline. However, existing methods typically perform dense computation over all images and spatial tokens, despite the fact that normal samples dominate real-world scenarios and anomalies usually occupy only small regions. Token pruning offers a promising solution, but introduces an asymmetric pruning risk in anomaly detection: retaining normal tokens mainly incurs redundant computation, whereas removing anomalous tokens may elim"},"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":"2608.03681","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-04T13:53:36Z","cross_cats_sorted":[],"title_canon_sha256":"f8d387b4a3bf93810715493655bff6a371a30085aee1364b3dd2716dde173ca3","abstract_canon_sha256":"a6fb11d4daab0bc6e0e10c09b269e24cedbe4b14315603e29bef3b3fcfe0d362"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-05T01:36:52.562788Z","signature_b64":"hzavHhjZfXw1DNAshyRwLAJX40x3sEFzEgbqVcyEEZG6ONq23DIEgE8Ni76MY0zmLB+4MlVvaSfuN6fKiFwgAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ed7a35b3be478876d30e29aa6ecc4fd3db9ad08162cf777bba49011442ccb39","last_reissued_at":"2026-08-05T01:36:52.561355Z","signature_status":"signed_v1","first_computed_at":"2026-08-05T01:36:52.561355Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Keep the Needle, Prune the Haystack: Defect-Preserving Token Pruning for Efficient Zero-Shot Anomaly Detection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jingyuan Zhang, Ke Xu, Qixiang Ma, Sihang Zhou, Xiaoyun Wang, Yanning Hou","submitted_at":"2026-08-04T13:53:36Z","abstract_excerpt":"Zero-shot visual anomaly detection has achieved remarkable progress, with recent vision-only approaches further improving performance while simplifying the inference pipeline. However, existing methods typically perform dense computation over all images and spatial tokens, despite the fact that normal samples dominate real-world scenarios and anomalies usually occupy only small regions. Token pruning offers a promising solution, but introduces an asymmetric pruning risk in anomaly detection: retaining normal tokens mainly incurs redundant computation, whereas removing anomalous tokens may elim"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.03681","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/2608.03681/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":"2608.03681","created_at":"2026-08-05T01:36:52.561793+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.03681v1","created_at":"2026-08-05T01:36:52.561793+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.03681","created_at":"2026-08-05T01:36:52.561793+00:00"},{"alias_kind":"pith_short_12","alias_value":"T3L2GWZ34R4I","created_at":"2026-08-05T01:36:52.561793+00:00"},{"alias_kind":"pith_short_16","alias_value":"T3L2GWZ34R4IO3JQ","created_at":"2026-08-05T01:36:52.561793+00:00"},{"alias_kind":"pith_short_8","alias_value":"T3L2GWZ3","created_at":"2026-08-05T01:36:52.561793+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/T3L2GWZ34R4IO3JQ4KNKN3GE7U","json":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U.json","graph_json":"https://pith.science/api/pith-number/T3L2GWZ34R4IO3JQ4KNKN3GE7U/graph.json","events_json":"https://pith.science/api/pith-number/T3L2GWZ34R4IO3JQ4KNKN3GE7U/events.json","paper":"https://pith.science/paper/T3L2GWZ3"},"agent_actions":{"view_html":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U","download_json":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U.json","view_paper":"https://pith.science/paper/T3L2GWZ3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.03681&json=true","fetch_graph":"https://pith.science/api/pith-number/T3L2GWZ34R4IO3JQ4KNKN3GE7U/graph.json","fetch_events":"https://pith.science/api/pith-number/T3L2GWZ34R4IO3JQ4KNKN3GE7U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U/action/storage_attestation","attest_author":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U/action/author_attestation","sign_citation":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U/action/citation_signature","submit_replication":"https://pith.science/pith/T3L2GWZ34R4IO3JQ4KNKN3GE7U/action/replication_record"}},"created_at":"2026-08-05T01:36:52.561793+00:00","updated_at":"2026-08-05T01:36:52.561793+00:00"}