{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:TSRBQIUVJGHVTE74QWZUIWFB33","short_pith_number":"pith:TSRBQIUV","canonical_record":{"source":{"id":"2502.09352","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T14:18:41Z","cross_cats_sorted":["cs.CV","math.OC"],"title_canon_sha256":"9da98f88c33e592ab781fb670f9eebdfb080c25c3b599f6ea95b711639bd931c","abstract_canon_sha256":"9ac62439c304c1e85cd302109600410215bc1407476d762da443a4b3e1f76088"},"schema_version":"1.0"},"canonical_sha256":"9ca2182295498f5993fc85b34458a1dec33102602044654bd55e8e2f1a69d4a3","source":{"kind":"arxiv","id":"2502.09352","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.09352","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"arxiv_version","alias_value":"2502.09352v1","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09352","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_12","alias_value":"TSRBQIUVJGHV","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_16","alias_value":"TSRBQIUVJGHVTE74","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_8","alias_value":"TSRBQIUV","created_at":"2026-07-05T10:13:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:TSRBQIUVJGHVTE74QWZUIWFB33","target":"record","payload":{"canonical_record":{"source":{"id":"2502.09352","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T14:18:41Z","cross_cats_sorted":["cs.CV","math.OC"],"title_canon_sha256":"9da98f88c33e592ab781fb670f9eebdfb080c25c3b599f6ea95b711639bd931c","abstract_canon_sha256":"9ac62439c304c1e85cd302109600410215bc1407476d762da443a4b3e1f76088"},"schema_version":"1.0"},"canonical_sha256":"9ca2182295498f5993fc85b34458a1dec33102602044654bd55e8e2f1a69d4a3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:13:56.235388Z","signature_b64":"FGBpyr4iJuup3Ahgg/rixwZGniGm4q9ymdHKK9X92OSp+GGY2rEhlBQz5fSP5dkIAyn4grwbu9KTBkWTn97XDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ca2182295498f5993fc85b34458a1dec33102602044654bd55e8e2f1a69d4a3","last_reissued_at":"2026-07-05T10:13:56.234882Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:13:56.234882Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.09352","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:13:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DIfSzpMWbkBYnxow8mQTyORw4lC8L8rM+gVOmlrCgMtLl8FizETRP/1JT9L0eGzhRShiPKzNyx50zfnszNYUCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:53:43.252715Z"},"content_sha256":"e704254ea8e6f1d702fadcb84515bd490c5a9897570c586b69bb7827ea7a422e","schema_version":"1.0","event_id":"sha256:e704254ea8e6f1d702fadcb84515bd490c5a9897570c586b69bb7827ea7a422e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:TSRBQIUVJGHVTE74QWZUIWFB33","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Wasserstein distributional adversarial training for deep neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV","math.OC"],"primary_cat":"cs.LG","authors_text":"Guangyi He, Jan Obloj, Xingjian Bai, Yifan Jiang","submitted_at":"2025-02-13T14:18:41Z","abstract_excerpt":"Design of adversarial attacks for deep neural networks, as well as methods of adversarial training against them, are subject of intense research. In this paper, we propose methods to train against distributional attack threats, extending the TRADES method used for pointwise attacks. Our approach leverages recent contributions and relies on sensitivity analysis for Wasserstein distributionally robust optimization problems. We introduce an efficient fine-tuning method which can be deployed on a previously trained model. We test our methods on a range of pre-trained models on RobustBench. These e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09352","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/2502.09352/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:13:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dgHP0uwYJ70TE5okun7Q6PQi6fQhjBY5cfc2Jf5bLBqGDcPSloKj7pEr0X3K+dlvdUGIFJlzH002DsJzP1AqDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T22:53:43.253655Z"},"content_sha256":"0b7c20f16ab4ddc21352e505afb61d0839dc1ec5c94ea25accfba546fb55a292","schema_version":"1.0","event_id":"sha256:0b7c20f16ab4ddc21352e505afb61d0839dc1ec5c94ea25accfba546fb55a292"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TSRBQIUVJGHVTE74QWZUIWFB33/bundle.json","state_url":"https://pith.science/pith/TSRBQIUVJGHVTE74QWZUIWFB33/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TSRBQIUVJGHVTE74QWZUIWFB33/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T22:53:43Z","links":{"resolver":"https://pith.science/pith/TSRBQIUVJGHVTE74QWZUIWFB33","bundle":"https://pith.science/pith/TSRBQIUVJGHVTE74QWZUIWFB33/bundle.json","state":"https://pith.science/pith/TSRBQIUVJGHVTE74QWZUIWFB33/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TSRBQIUVJGHVTE74QWZUIWFB33/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:TSRBQIUVJGHVTE74QWZUIWFB33","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9ac62439c304c1e85cd302109600410215bc1407476d762da443a4b3e1f76088","cross_cats_sorted":["cs.CV","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T14:18:41Z","title_canon_sha256":"9da98f88c33e592ab781fb670f9eebdfb080c25c3b599f6ea95b711639bd931c"},"schema_version":"1.0","source":{"id":"2502.09352","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.09352","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"arxiv_version","alias_value":"2502.09352v1","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.09352","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_12","alias_value":"TSRBQIUVJGHV","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_16","alias_value":"TSRBQIUVJGHVTE74","created_at":"2026-07-05T10:13:56Z"},{"alias_kind":"pith_short_8","alias_value":"TSRBQIUV","created_at":"2026-07-05T10:13:56Z"}],"graph_snapshots":[{"event_id":"sha256:0b7c20f16ab4ddc21352e505afb61d0839dc1ec5c94ea25accfba546fb55a292","target":"graph","created_at":"2026-07-05T10:13:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2502.09352/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Design of adversarial attacks for deep neural networks, as well as methods of adversarial training against them, are subject of intense research. In this paper, we propose methods to train against distributional attack threats, extending the TRADES method used for pointwise attacks. Our approach leverages recent contributions and relies on sensitivity analysis for Wasserstein distributionally robust optimization problems. We introduce an efficient fine-tuning method which can be deployed on a previously trained model. We test our methods on a range of pre-trained models on RobustBench. These e","authors_text":"Guangyi He, Jan Obloj, Xingjian Bai, Yifan Jiang","cross_cats":["cs.CV","math.OC"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T14:18:41Z","title":"Wasserstein distributional adversarial training for deep neural networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.09352","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e704254ea8e6f1d702fadcb84515bd490c5a9897570c586b69bb7827ea7a422e","target":"record","created_at":"2026-07-05T10:13:56Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"9ac62439c304c1e85cd302109600410215bc1407476d762da443a4b3e1f76088","cross_cats_sorted":["cs.CV","math.OC"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-13T14:18:41Z","title_canon_sha256":"9da98f88c33e592ab781fb670f9eebdfb080c25c3b599f6ea95b711639bd931c"},"schema_version":"1.0","source":{"id":"2502.09352","kind":"arxiv","version":1}},"canonical_sha256":"9ca2182295498f5993fc85b34458a1dec33102602044654bd55e8e2f1a69d4a3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9ca2182295498f5993fc85b34458a1dec33102602044654bd55e8e2f1a69d4a3","first_computed_at":"2026-07-05T10:13:56.234882Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:13:56.234882Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FGBpyr4iJuup3Ahgg/rixwZGniGm4q9ymdHKK9X92OSp+GGY2rEhlBQz5fSP5dkIAyn4grwbu9KTBkWTn97XDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:13:56.235388Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.09352","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e704254ea8e6f1d702fadcb84515bd490c5a9897570c586b69bb7827ea7a422e","sha256:0b7c20f16ab4ddc21352e505afb61d0839dc1ec5c94ea25accfba546fb55a292"],"state_sha256":"e22b54dd4aee09bb8431aa726657f9754b0f9c59ef0b4f09d32f189b9157183b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p2sR5Zy0nwEz61KLZTc+S25V0Jhg34wpAfSFcNhxMYFtjTKqcuN9nRoHZ+F+3IVRfB0+3+gmr32xqxwVMJunDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T22:53:43.261647Z","bundle_sha256":"6fd74571252f85d13001dce573500ef7c9f5ec5ebde47fccc09cacfaceba1fb2"}}