{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RHHDNDNNYBZ3AU5B7AXKKKZSID","short_pith_number":"pith:RHHDNDNN","schema_version":"1.0","canonical_sha256":"89ce368dadc073b053a1f82ea52b3240f18eeb321429219a114ee2a62a6df4f9","source":{"kind":"arxiv","id":"2208.05740","version":2},"attestation_state":"computed","paper":{"title":"General Cutting Planes for Bound-Propagation-Based Neural Network Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CV","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Bo Li, Cho-Jui Hsieh, Huan Zhang, J. Zico Kolter, Kaidi Xu, Linyi Li, Shiqi Wang, Suman Jana","submitted_at":"2022-08-11T10:31:28Z","abstract_excerpt":"Bound propagation methods, when combined with branch and bound, are among the most effective methods to formally verify properties of deep neural networks such as correctness, robustness, and safety. However, existing works cannot handle the general form of cutting plane constraints widely accepted in traditional solvers, which are crucial for strengthening verifiers with tightened convex relaxations. In this paper, we generalize the bound propagation procedure to allow the addition of arbitrary cutting plane constraints, including those involving relaxed integer variables that do not appear i"},"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":"2208.05740","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-08-11T10:31:28Z","cross_cats_sorted":["cs.CR","cs.CV","math.OC","stat.ML"],"title_canon_sha256":"6c93f8703b24459c81475721d4de70e2bb7cfff1a5435cc993b3323bb0632316","abstract_canon_sha256":"5492c8595d09c8b169ce857d40d02430cc80d8e7b2ba2ad248bd2fc16ad0ef40"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:22:09.851475Z","signature_b64":"IxUbRib5ZH241KY/PYQG/fJX2Fykzu+no/aRfMWUpoJsO6N1VL9g3ehsRqqUAuFpNdKiU3mDT4RNhFLGMLmCCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89ce368dadc073b053a1f82ea52b3240f18eeb321429219a114ee2a62a6df4f9","last_reissued_at":"2026-07-05T05:22:09.850943Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:22:09.850943Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"General Cutting Planes for Bound-Propagation-Based Neural Network Verification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CR","cs.CV","math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Bo Li, Cho-Jui Hsieh, Huan Zhang, J. Zico Kolter, Kaidi Xu, Linyi Li, Shiqi Wang, Suman Jana","submitted_at":"2022-08-11T10:31:28Z","abstract_excerpt":"Bound propagation methods, when combined with branch and bound, are among the most effective methods to formally verify properties of deep neural networks such as correctness, robustness, and safety. However, existing works cannot handle the general form of cutting plane constraints widely accepted in traditional solvers, which are crucial for strengthening verifiers with tightened convex relaxations. In this paper, we generalize the bound propagation procedure to allow the addition of arbitrary cutting plane constraints, including those involving relaxed integer variables that do not appear i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.05740","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/2208.05740/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":"2208.05740","created_at":"2026-07-05T05:22:09.851004+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.05740v2","created_at":"2026-07-05T05:22:09.851004+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.05740","created_at":"2026-07-05T05:22:09.851004+00:00"},{"alias_kind":"pith_short_12","alias_value":"RHHDNDNNYBZ3","created_at":"2026-07-05T05:22:09.851004+00:00"},{"alias_kind":"pith_short_16","alias_value":"RHHDNDNNYBZ3AU5B","created_at":"2026-07-05T05:22:09.851004+00:00"},{"alias_kind":"pith_short_8","alias_value":"RHHDNDNN","created_at":"2026-07-05T05:22:09.851004+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2411.18235","citing_title":"Certified Training with Branch-and-Bound for Lyapunov-stable Neural Control","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13845","citing_title":"Quantitative Linear Logic for Neuro-Symbolic Learning and Verification","ref_index":134,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13845","citing_title":"Quantitative Linear Logic for Neuro-Symbolic Learning and Verification","ref_index":134,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID","json":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID.json","graph_json":"https://pith.science/api/pith-number/RHHDNDNNYBZ3AU5B7AXKKKZSID/graph.json","events_json":"https://pith.science/api/pith-number/RHHDNDNNYBZ3AU5B7AXKKKZSID/events.json","paper":"https://pith.science/paper/RHHDNDNN"},"agent_actions":{"view_html":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID","download_json":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID.json","view_paper":"https://pith.science/paper/RHHDNDNN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.05740&json=true","fetch_graph":"https://pith.science/api/pith-number/RHHDNDNNYBZ3AU5B7AXKKKZSID/graph.json","fetch_events":"https://pith.science/api/pith-number/RHHDNDNNYBZ3AU5B7AXKKKZSID/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID/action/storage_attestation","attest_author":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID/action/author_attestation","sign_citation":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID/action/citation_signature","submit_replication":"https://pith.science/pith/RHHDNDNNYBZ3AU5B7AXKKKZSID/action/replication_record"}},"created_at":"2026-07-05T05:22:09.851004+00:00","updated_at":"2026-07-05T05:22:09.851004+00:00"}