{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:MSEC7IG7ZN72ZNXFDSPQJTU7RY","short_pith_number":"pith:MSEC7IG7","schema_version":"1.0","canonical_sha256":"64882fa0dfcb7facb6e51c9f04ce9f8e26b8646d626fee55300657ff7447d976","source":{"kind":"arxiv","id":"2608.09047","version":1},"attestation_state":"computed","paper":{"title":"Diversity Matters: Distributional Feature Coverage Sample Selection for Data-Efficient Backdoor Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CR","authors_text":"Feng-Qi Cui, Haiming Jin, Jia-Cheng Zhao, Jinyang Huang, Meng Li, Xiaokang Zhou, Xiaoke Chen, Yi Yang, Yu-Tong Guo","submitted_at":"2026-08-10T02:52:24Z","abstract_excerpt":"Backdoor attacks compromise training data so that a model retains clean accuracy but predicts an attacker-chosen target on triggered inputs. At very low poisoning rates, only a few samples convey the trigger--target association, making poison-sample selection critical. Existing methods typically rank candidates using per-sample scores, which can select redundant samples from similar semantic regions, and many require task-specific surrogate training. We propose Distributional Feature Coverage Sample Selection (DFCS), a training-free, trigger-agnostic method that clusters fixed pretrained featu"},"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":"2608.09047","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2026-08-10T02:52:24Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b49d16384b53802455a3c2d769c7c39c10975e21877daf0508854322798b58a0","abstract_canon_sha256":"c27a59402603b54f05bbecf4f68a9b17be97969724394d6c7158b8ce16cd3ef8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T02:21:40.851964Z","signature_b64":"jxYLT8Rq4asmBN/yTcz2o8Udk6NmeTgQn3++A/Fnrq9QpBIWPnHlvCpM9PusX1MOiemJ441v+doP6xQUoCtmAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"64882fa0dfcb7facb6e51c9f04ce9f8e26b8646d626fee55300657ff7447d976","last_reissued_at":"2026-08-11T02:21:40.849930Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T02:21:40.849930Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diversity Matters: Distributional Feature Coverage Sample Selection for Data-Efficient Backdoor Attacks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CR","authors_text":"Feng-Qi Cui, Haiming Jin, Jia-Cheng Zhao, Jinyang Huang, Meng Li, Xiaokang Zhou, Xiaoke Chen, Yi Yang, Yu-Tong Guo","submitted_at":"2026-08-10T02:52:24Z","abstract_excerpt":"Backdoor attacks compromise training data so that a model retains clean accuracy but predicts an attacker-chosen target on triggered inputs. At very low poisoning rates, only a few samples convey the trigger--target association, making poison-sample selection critical. Existing methods typically rank candidates using per-sample scores, which can select redundant samples from similar semantic regions, and many require task-specific surrogate training. We propose Distributional Feature Coverage Sample Selection (DFCS), a training-free, trigger-agnostic method that clusters fixed pretrained featu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09047","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/2608.09047/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":"2608.09047","created_at":"2026-08-11T02:21:40.850816+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.09047v1","created_at":"2026-08-11T02:21:40.850816+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09047","created_at":"2026-08-11T02:21:40.850816+00:00"},{"alias_kind":"pith_short_12","alias_value":"MSEC7IG7ZN72","created_at":"2026-08-11T02:21:40.850816+00:00"},{"alias_kind":"pith_short_16","alias_value":"MSEC7IG7ZN72ZNXF","created_at":"2026-08-11T02:21:40.850816+00:00"},{"alias_kind":"pith_short_8","alias_value":"MSEC7IG7","created_at":"2026-08-11T02:21:40.850816+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/MSEC7IG7ZN72ZNXFDSPQJTU7RY","json":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY.json","graph_json":"https://pith.science/api/pith-number/MSEC7IG7ZN72ZNXFDSPQJTU7RY/graph.json","events_json":"https://pith.science/api/pith-number/MSEC7IG7ZN72ZNXFDSPQJTU7RY/events.json","paper":"https://pith.science/paper/MSEC7IG7"},"agent_actions":{"view_html":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY","download_json":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY.json","view_paper":"https://pith.science/paper/MSEC7IG7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.09047&json=true","fetch_graph":"https://pith.science/api/pith-number/MSEC7IG7ZN72ZNXFDSPQJTU7RY/graph.json","fetch_events":"https://pith.science/api/pith-number/MSEC7IG7ZN72ZNXFDSPQJTU7RY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY/action/storage_attestation","attest_author":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY/action/author_attestation","sign_citation":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY/action/citation_signature","submit_replication":"https://pith.science/pith/MSEC7IG7ZN72ZNXFDSPQJTU7RY/action/replication_record"}},"created_at":"2026-08-11T02:21:40.850816+00:00","updated_at":"2026-08-11T02:21:40.850816+00:00"}