{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:4NC5YNAIF6ZD3SSX34HYWFEEZW","short_pith_number":"pith:4NC5YNAI","canonical_record":{"source":{"id":"2201.02009","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-06T11:00:49Z","cross_cats_sorted":[],"title_canon_sha256":"0ec5fdf34f3947a46aa4e853711b9c40aefd9f20efda2d0fa28fcc58a52bdd4b","abstract_canon_sha256":"5a47825f4426d41ebbbdb709b0ae06d2573b361eb448d2cc32132158bff61527"},"schema_version":"1.0"},"canonical_sha256":"e345dc34082fb23dca57df0f8b1484cd87529a09c030f09f6f3b031ac58c0206","source":{"kind":"arxiv","id":"2201.02009","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.02009","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"arxiv_version","alias_value":"2201.02009v2","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.02009","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_12","alias_value":"4NC5YNAIF6ZD","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_16","alias_value":"4NC5YNAIF6ZD3SSX","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_8","alias_value":"4NC5YNAI","created_at":"2026-07-05T05:09:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:4NC5YNAIF6ZD3SSX34HYWFEEZW","target":"record","payload":{"canonical_record":{"source":{"id":"2201.02009","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-06T11:00:49Z","cross_cats_sorted":[],"title_canon_sha256":"0ec5fdf34f3947a46aa4e853711b9c40aefd9f20efda2d0fa28fcc58a52bdd4b","abstract_canon_sha256":"5a47825f4426d41ebbbdb709b0ae06d2573b361eb448d2cc32132158bff61527"},"schema_version":"1.0"},"canonical_sha256":"e345dc34082fb23dca57df0f8b1484cd87529a09c030f09f6f3b031ac58c0206","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:09:14.723118Z","signature_b64":"lx6jtlW1BBr6KB1iaHJJUcCtSVdMk+cY40tp74tKMtG1PxlovmAxN50jmK1a3LXCpglRDci5hJo/+3s+w87MAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e345dc34082fb23dca57df0f8b1484cd87529a09c030f09f6f3b031ac58c0206","last_reissued_at":"2026-07-05T05:09:14.722734Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:09:14.722734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2201.02009","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:09:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OphCXqQJ/HWe/+rQzPIBrDcLnfBhocDzEijiOeB8+xjV3czsjuyGGrGOq/tgINq+6nFT45sSpp6buzqG6x2uDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T10:04:06.241206Z"},"content_sha256":"eab00a0f4cc032ddc22995667bbb297b685f02d13763bd7e8e4af9acbec2900d","schema_version":"1.0","event_id":"sha256:eab00a0f4cc032ddc22995667bbb297b685f02d13763bd7e8e4af9acbec2900d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:4NC5YNAIF6ZD3SSX34HYWFEEZW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dongdong Zhang, Jian Yang, Juncheng Wan, Shuming Ma, Weinan Zhang, Yong Yu, Zhoujun Li","submitted_at":"2022-01-06T11:00:49Z","abstract_excerpt":"While end-to-end neural machine translation (NMT) has achieved impressive progress, noisy input usually leads models to become fragile and unstable. Generating adversarial examples as the augmented data has been proved to be useful to alleviate this problem. Existing methods for adversarial example generation (AEG) are word-level or character-level, which ignore the ubiquitous phrase structure. In this paper, we propose a Phrase-level Adversarial Example Generation (PAEG) framework to enhance the robustness of the translation model. Our method further improves the gradient-based word-level AEG"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.02009","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/2201.02009/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":"2026-07-05T05:09:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zx5CWs/ztHIrYlW1PPDKGXH5Yp/pgJhmVaCq1OFkCOpRR+nsFT7DPKhFTUlYE8fZ7flu7Vsn+FQ3FOvUSlMaCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T10:04:06.242790Z"},"content_sha256":"69751e1264c4d0eb354afb8b037628fd2f3f2325985f47b48e1539de4924fb0a","schema_version":"1.0","event_id":"sha256:69751e1264c4d0eb354afb8b037628fd2f3f2325985f47b48e1539de4924fb0a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/bundle.json","state_url":"https://pith.science/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/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":"2026-08-01T10:04:06Z","links":{"resolver":"https://pith.science/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW","bundle":"https://pith.science/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/bundle.json","state":"https://pith.science/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4NC5YNAIF6ZD3SSX34HYWFEEZW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:4NC5YNAIF6ZD3SSX34HYWFEEZW","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":"5a47825f4426d41ebbbdb709b0ae06d2573b361eb448d2cc32132158bff61527","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-06T11:00:49Z","title_canon_sha256":"0ec5fdf34f3947a46aa4e853711b9c40aefd9f20efda2d0fa28fcc58a52bdd4b"},"schema_version":"1.0","source":{"id":"2201.02009","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.02009","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"arxiv_version","alias_value":"2201.02009v2","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.02009","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_12","alias_value":"4NC5YNAIF6ZD","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_16","alias_value":"4NC5YNAIF6ZD3SSX","created_at":"2026-07-05T05:09:14Z"},{"alias_kind":"pith_short_8","alias_value":"4NC5YNAI","created_at":"2026-07-05T05:09:14Z"}],"graph_snapshots":[{"event_id":"sha256:69751e1264c4d0eb354afb8b037628fd2f3f2325985f47b48e1539de4924fb0a","target":"graph","created_at":"2026-07-05T05:09:14Z","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/2201.02009/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While end-to-end neural machine translation (NMT) has achieved impressive progress, noisy input usually leads models to become fragile and unstable. Generating adversarial examples as the augmented data has been proved to be useful to alleviate this problem. Existing methods for adversarial example generation (AEG) are word-level or character-level, which ignore the ubiquitous phrase structure. In this paper, we propose a Phrase-level Adversarial Example Generation (PAEG) framework to enhance the robustness of the translation model. Our method further improves the gradient-based word-level AEG","authors_text":"Dongdong Zhang, Jian Yang, Juncheng Wan, Shuming Ma, Weinan Zhang, Yong Yu, Zhoujun Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-06T11:00:49Z","title":"PAEG: Phrase-level Adversarial Example Generation for Neural Machine Translation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.02009","kind":"arxiv","version":2},"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:eab00a0f4cc032ddc22995667bbb297b685f02d13763bd7e8e4af9acbec2900d","target":"record","created_at":"2026-07-05T05:09:14Z","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":"5a47825f4426d41ebbbdb709b0ae06d2573b361eb448d2cc32132158bff61527","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-06T11:00:49Z","title_canon_sha256":"0ec5fdf34f3947a46aa4e853711b9c40aefd9f20efda2d0fa28fcc58a52bdd4b"},"schema_version":"1.0","source":{"id":"2201.02009","kind":"arxiv","version":2}},"canonical_sha256":"e345dc34082fb23dca57df0f8b1484cd87529a09c030f09f6f3b031ac58c0206","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e345dc34082fb23dca57df0f8b1484cd87529a09c030f09f6f3b031ac58c0206","first_computed_at":"2026-07-05T05:09:14.722734Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:09:14.722734Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lx6jtlW1BBr6KB1iaHJJUcCtSVdMk+cY40tp74tKMtG1PxlovmAxN50jmK1a3LXCpglRDci5hJo/+3s+w87MAw==","signature_status":"signed_v1","signed_at":"2026-07-05T05:09:14.723118Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.02009","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:eab00a0f4cc032ddc22995667bbb297b685f02d13763bd7e8e4af9acbec2900d","sha256:69751e1264c4d0eb354afb8b037628fd2f3f2325985f47b48e1539de4924fb0a"],"state_sha256":"0c261d30ffa2a46a5e37d77e4d8dde21823025e0079190eaa40c181ff0d15f20"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sg9KxuwqI7iu9Bew2RT353M1EHJYG0AATqxQCgRzurFTxqx2zGCnV730DaSuFYnxs4AFXTPgxDR5cBpUtGawDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T10:04:06.247476Z","bundle_sha256":"b6bbb2772f7f06aa69702d465c69bca0d267ae8d84b2e0e14e286bfe02e6cded"}}