{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:KXSAER56ETE2VXQKRL74LHEUKG","short_pith_number":"pith:KXSAER56","schema_version":"1.0","canonical_sha256":"55e40247be24c9aade0a8affc59c9451830dd5bb1327068a405568a2479f77e6","source":{"kind":"arxiv","id":"2410.08190","version":2},"attestation_state":"computed","paper":{"title":"Poison-splat: Computation Cost Attack on 3D Gaussian Splatting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiahao Lu, Qiuhong Shen, Shuicheng Yan, Xinchao Wang, Yifan Zhang","submitted_at":"2024-10-10T17:57:29Z","abstract_excerpt":"3D Gaussian splatting (3DGS), known for its groundbreaking performance and efficiency, has become a dominant 3D representation and brought progress to many 3D vision tasks. However, in this work, we reveal a significant security vulnerability that has been largely overlooked in 3DGS: the computation cost of training 3DGS could be maliciously tampered by poisoning the input data. By developing an attack named Poison-splat, we reveal a novel attack surface where the adversary can poison the input images to drastically increase the computation memory and time needed for 3DGS training, pushing the"},"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":"2410.08190","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-10T17:57:29Z","cross_cats_sorted":["cs.CR","cs.GR","cs.LG"],"title_canon_sha256":"f9543ee29ed5bc0f589b98a6c847a562b746564acb3408d1a66d2cfb1d4628ed","abstract_canon_sha256":"8c731a744832c31d65247f9c7a13cf52daf1689a0346929efafe329b42ebcd2d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:22:51.963001Z","signature_b64":"om7VmA+1qcDysVQ6PBDnAdTTs4EP4EeIMD8KppbsjI94aLwtKnl1ZRUgnVsneE5874IwxtifE3ws/D8I8RH3DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55e40247be24c9aade0a8affc59c9451830dd5bb1327068a405568a2479f77e6","last_reissued_at":"2026-07-05T10:22:51.962311Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:22:51.962311Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Poison-splat: Computation Cost Attack on 3D Gaussian Splatting","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.GR","cs.LG"],"primary_cat":"cs.CV","authors_text":"Jiahao Lu, Qiuhong Shen, Shuicheng Yan, Xinchao Wang, Yifan Zhang","submitted_at":"2024-10-10T17:57:29Z","abstract_excerpt":"3D Gaussian splatting (3DGS), known for its groundbreaking performance and efficiency, has become a dominant 3D representation and brought progress to many 3D vision tasks. However, in this work, we reveal a significant security vulnerability that has been largely overlooked in 3DGS: the computation cost of training 3DGS could be maliciously tampered by poisoning the input data. By developing an attack named Poison-splat, we reveal a novel attack surface where the adversary can poison the input images to drastically increase the computation memory and time needed for 3DGS training, pushing the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.08190","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/2410.08190/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":"2410.08190","created_at":"2026-07-05T10:22:51.962389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.08190v2","created_at":"2026-07-05T10:22:51.962389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.08190","created_at":"2026-07-05T10:22:51.962389+00:00"},{"alias_kind":"pith_short_12","alias_value":"KXSAER56ETE2","created_at":"2026-07-05T10:22:51.962389+00:00"},{"alias_kind":"pith_short_16","alias_value":"KXSAER56ETE2VXQK","created_at":"2026-07-05T10:22:51.962389+00:00"},{"alias_kind":"pith_short_8","alias_value":"KXSAER56","created_at":"2026-07-05T10:22:51.962389+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03499","citing_title":"Characterizing Detectability in 3DGS Poisoning: A Stage-wise Benchmark","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00450","citing_title":"Optimizing 3D Gaussian Splatting via Point Cloud Upsampling","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG","json":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG.json","graph_json":"https://pith.science/api/pith-number/KXSAER56ETE2VXQKRL74LHEUKG/graph.json","events_json":"https://pith.science/api/pith-number/KXSAER56ETE2VXQKRL74LHEUKG/events.json","paper":"https://pith.science/paper/KXSAER56"},"agent_actions":{"view_html":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG","download_json":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG.json","view_paper":"https://pith.science/paper/KXSAER56","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.08190&json=true","fetch_graph":"https://pith.science/api/pith-number/KXSAER56ETE2VXQKRL74LHEUKG/graph.json","fetch_events":"https://pith.science/api/pith-number/KXSAER56ETE2VXQKRL74LHEUKG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG/action/storage_attestation","attest_author":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG/action/author_attestation","sign_citation":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG/action/citation_signature","submit_replication":"https://pith.science/pith/KXSAER56ETE2VXQKRL74LHEUKG/action/replication_record"}},"created_at":"2026-07-05T10:22:51.962389+00:00","updated_at":"2026-07-05T10:22:51.962389+00:00"}