{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:SQ7MBV77PMVOQHOOHKSR5B77R7","short_pith_number":"pith:SQ7MBV77","canonical_record":{"source":{"id":"2103.07352","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-12T15:31:34Z","cross_cats_sorted":[],"title_canon_sha256":"14ed6003d4bf2b790dbf45c178d6e5c1e4eda631ab71040b606308c6794f7d56","abstract_canon_sha256":"0a3a5b563295a61257e60d7f32829183a93c5da4b0794cc7176a67a7f1444d7e"},"schema_version":"1.0"},"canonical_sha256":"943ec0d7ff7b2ae81dce3aa51e87ff8ff33eb7c0b49740e9c24155b8a77af1d6","source":{"kind":"arxiv","id":"2103.07352","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.07352","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"arxiv_version","alias_value":"2103.07352v3","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.07352","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_12","alias_value":"SQ7MBV77PMVO","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_16","alias_value":"SQ7MBV77PMVOQHOO","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_8","alias_value":"SQ7MBV77","created_at":"2026-07-05T04:19:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:SQ7MBV77PMVOQHOOHKSR5B77R7","target":"record","payload":{"canonical_record":{"source":{"id":"2103.07352","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-12T15:31:34Z","cross_cats_sorted":[],"title_canon_sha256":"14ed6003d4bf2b790dbf45c178d6e5c1e4eda631ab71040b606308c6794f7d56","abstract_canon_sha256":"0a3a5b563295a61257e60d7f32829183a93c5da4b0794cc7176a67a7f1444d7e"},"schema_version":"1.0"},"canonical_sha256":"943ec0d7ff7b2ae81dce3aa51e87ff8ff33eb7c0b49740e9c24155b8a77af1d6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:19:21.726251Z","signature_b64":"lQKuQ5xajwaTe0mfgy/fFpDDuwnQ2U/nZjBK7Uu4ztpMYzeLPs3pHNnxTyb6j7hZXVqzPNeum/pt22cHzTBVBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"943ec0d7ff7b2ae81dce3aa51e87ff8ff33eb7c0b49740e9c24155b8a77af1d6","last_reissued_at":"2026-07-05T04:19:21.725784Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:19:21.725784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2103.07352","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:19:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rX466CUnoKuapVkWleyhzHspsafU1ey7ppaBqh1BIcXzvnvavALX1G/EKAwRQYOXRNsn3fW2OinNk2nKMzhTAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:26:37.170914Z"},"content_sha256":"7a0978e7e304282800d34119549aa2233c3ec4be3fd49a0d86c80d11e90d176e","schema_version":"1.0","event_id":"sha256:7a0978e7e304282800d34119549aa2233c3ec4be3fd49a0d86c80d11e90d176e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:SQ7MBV77PMVOQHOOHKSR5B77R7","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Visual Cues and Error Correction for Translation Robustness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Lucia Specia, Marek Rei, Zhenhao Li","submitted_at":"2021-03-12T15:31:34Z","abstract_excerpt":"Neural Machine Translation models are sensitive to noise in the input texts, such as misspelled words and ungrammatical constructions. Existing robustness techniques generally fail when faced with unseen types of noise and their performance degrades on clean texts. In this paper, we focus on three types of realistic noise that are commonly generated by humans and introduce the idea of visual context to improve translation robustness for noisy texts. In addition, we describe a novel error correction training regime that can be used as an auxiliary task to further improve translation robustness."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.07352","kind":"arxiv","version":3},"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/2103.07352/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":"2026-07-05T04:19:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CjyC9VKsCz5dtWd9Nxq5pPsAaoLKKjIRkYpWDW8d6+2zCLEPNvgBoSwjYynwZm5PIfy78j3UPtTg8MKYLl3UAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:26:37.171420Z"},"content_sha256":"0196c8c37fef7513af9491516210d7089b44d7cc0bb5546c306684c92f562dfd","schema_version":"1.0","event_id":"sha256:0196c8c37fef7513af9491516210d7089b44d7cc0bb5546c306684c92f562dfd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/bundle.json","state_url":"https://pith.science/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/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":"2026-08-03T16:26:37Z","links":{"resolver":"https://pith.science/pith/SQ7MBV77PMVOQHOOHKSR5B77R7","bundle":"https://pith.science/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/bundle.json","state":"https://pith.science/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SQ7MBV77PMVOQHOOHKSR5B77R7/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:SQ7MBV77PMVOQHOOHKSR5B77R7","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":"0a3a5b563295a61257e60d7f32829183a93c5da4b0794cc7176a67a7f1444d7e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-12T15:31:34Z","title_canon_sha256":"14ed6003d4bf2b790dbf45c178d6e5c1e4eda631ab71040b606308c6794f7d56"},"schema_version":"1.0","source":{"id":"2103.07352","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2103.07352","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"arxiv_version","alias_value":"2103.07352v3","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.07352","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_12","alias_value":"SQ7MBV77PMVO","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_16","alias_value":"SQ7MBV77PMVOQHOO","created_at":"2026-07-05T04:19:21Z"},{"alias_kind":"pith_short_8","alias_value":"SQ7MBV77","created_at":"2026-07-05T04:19:21Z"}],"graph_snapshots":[{"event_id":"sha256:0196c8c37fef7513af9491516210d7089b44d7cc0bb5546c306684c92f562dfd","target":"graph","created_at":"2026-07-05T04:19:21Z","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/2103.07352/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural Machine Translation models are sensitive to noise in the input texts, such as misspelled words and ungrammatical constructions. Existing robustness techniques generally fail when faced with unseen types of noise and their performance degrades on clean texts. In this paper, we focus on three types of realistic noise that are commonly generated by humans and introduce the idea of visual context to improve translation robustness for noisy texts. In addition, we describe a novel error correction training regime that can be used as an auxiliary task to further improve translation robustness.","authors_text":"Lucia Specia, Marek Rei, Zhenhao Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-12T15:31:34Z","title":"Visual Cues and Error Correction for Translation Robustness"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.07352","kind":"arxiv","version":3},"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":"sha256:7a0978e7e304282800d34119549aa2233c3ec4be3fd49a0d86c80d11e90d176e","target":"record","created_at":"2026-07-05T04:19:21Z","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":"0a3a5b563295a61257e60d7f32829183a93c5da4b0794cc7176a67a7f1444d7e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-03-12T15:31:34Z","title_canon_sha256":"14ed6003d4bf2b790dbf45c178d6e5c1e4eda631ab71040b606308c6794f7d56"},"schema_version":"1.0","source":{"id":"2103.07352","kind":"arxiv","version":3}},"canonical_sha256":"943ec0d7ff7b2ae81dce3aa51e87ff8ff33eb7c0b49740e9c24155b8a77af1d6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"943ec0d7ff7b2ae81dce3aa51e87ff8ff33eb7c0b49740e9c24155b8a77af1d6","first_computed_at":"2026-07-05T04:19:21.725784Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:19:21.725784Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lQKuQ5xajwaTe0mfgy/fFpDDuwnQ2U/nZjBK7Uu4ztpMYzeLPs3pHNnxTyb6j7hZXVqzPNeum/pt22cHzTBVBg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:19:21.726251Z","signed_message":"canonical_sha256_bytes"},"source_id":"2103.07352","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7a0978e7e304282800d34119549aa2233c3ec4be3fd49a0d86c80d11e90d176e","sha256:0196c8c37fef7513af9491516210d7089b44d7cc0bb5546c306684c92f562dfd"],"state_sha256":"9e7114a73df017b1af72a4592c46824f9bf6faffbe4274f7d148aafef0341e44"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0NYIIUlutk2Pl+S1CRGsLkggS3wniYGPn93ark/RGncYuYB9am4xQ7KfIeH5/8l7zT255gVByWExwZxTz3LYDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:26:37.176206Z","bundle_sha256":"bd214bbde8757ee30cc015e74ea5a5079deac4a7134a09b7e76a554d6d733afb"}}