{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GJYE2Z74N4TD56ICL65XVTAYU5","short_pith_number":"pith:GJYE2Z74","canonical_record":{"source":{"id":"2401.16247","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T15:49:40Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"65d819b639cfe0c88bab6c08fb06f53296c6a8b892af95deda89f64f00b5c56e","abstract_canon_sha256":"13d371846b5301c56c4e6ec8e29601e9978eeac28bbe01a1b9a22b64762fa76e"},"schema_version":"1.0"},"canonical_sha256":"32704d67fc6f263ef9025fbb7acc18a77f922fcf7a62f288f7545aed52627d5f","source":{"kind":"arxiv","id":"2401.16247","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.16247","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2401.16247v1","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16247","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"GJYE2Z74N4TD","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_16","alias_value":"GJYE2Z74N4TD56IC","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_8","alias_value":"GJYE2Z74","created_at":"2026-07-05T07:38:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GJYE2Z74N4TD56ICL65XVTAYU5","target":"record","payload":{"canonical_record":{"source":{"id":"2401.16247","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T15:49:40Z","cross_cats_sorted":["cs.CY"],"title_canon_sha256":"65d819b639cfe0c88bab6c08fb06f53296c6a8b892af95deda89f64f00b5c56e","abstract_canon_sha256":"13d371846b5301c56c4e6ec8e29601e9978eeac28bbe01a1b9a22b64762fa76e"},"schema_version":"1.0"},"canonical_sha256":"32704d67fc6f263ef9025fbb7acc18a77f922fcf7a62f288f7545aed52627d5f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:38:46.597540Z","signature_b64":"odffuCElOqdylB/UZNymJhfhKkSJRRuDcmmpxxHEDPrWbqhwZ5vIUNQYKjN5c+HQydgriT+NMsZdv6HGvYcKBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"32704d67fc6f263ef9025fbb7acc18a77f922fcf7a62f288f7545aed52627d5f","last_reissued_at":"2026-07-05T07:38:46.597096Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:38:46.597096Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.16247","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":"2026-07-05T07:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rayqCe9Z94GN8z9rtEGbE886XWGrdUPkCNPnHEYv74cjatFtdSVQay+yD/ITzBTKlv4x0WXVOb2lGXUuzBCJDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T20:09:09.272761Z"},"content_sha256":"ecd508fc86cb694cde16dda343580f6f425da9bda30ca28f60db66df2f9e4209","schema_version":"1.0","event_id":"sha256:ecd508fc86cb694cde16dda343580f6f425da9bda30ca28f60db66df2f9e4209"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GJYE2Z74N4TD56ICL65XVTAYU5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Red Teaming in Multimodal and Multilingual Translation","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.CY"],"primary_cat":"cs.CL","authors_text":"Christophe Ropers, Christophe Touret, Corinne Wong, Cristian Canton Ferrer, David Dale, Gabriel Mejia Gonzalez, Ivan Evtimov, Kristina Pereyra, Marta R. Costa-juss\\`a, Pierre Andrews, Prangthip Hansanti, Seohyun Sonia Kim","submitted_at":"2024-01-29T15:49:40Z","abstract_excerpt":"Assessing performance in Natural Language Processing is becoming increasingly complex. One particular challenge is the potential for evaluation datasets to overlap with training data, either directly or indirectly, which can lead to skewed results and overestimation of model performance. As a consequence, human evaluation is gaining increasing interest as a means to assess the performance and reliability of models. One such method is the red teaming approach, which aims to generate edge cases where a model will produce critical errors. While this methodology is becoming standard practice for g"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16247","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/2401.16247/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":"2026-07-05T07:38:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zrywud8ajjd195D7l9X5OgTQMf4b88FElwd/BLLVJU8Orv3ZbSHECRVsCfg9g3RYbpArI1RBkvQMzNg5WjZnDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T20:09:09.273133Z"},"content_sha256":"3587c9e56af949d8232400fa8256006235ed4f5a68262de2470f51589d582b45","schema_version":"1.0","event_id":"sha256:3587c9e56af949d8232400fa8256006235ed4f5a68262de2470f51589d582b45"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GJYE2Z74N4TD56ICL65XVTAYU5/bundle.json","state_url":"https://pith.science/pith/GJYE2Z74N4TD56ICL65XVTAYU5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GJYE2Z74N4TD56ICL65XVTAYU5/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":"2026-08-22T20:09:09Z","links":{"resolver":"https://pith.science/pith/GJYE2Z74N4TD56ICL65XVTAYU5","bundle":"https://pith.science/pith/GJYE2Z74N4TD56ICL65XVTAYU5/bundle.json","state":"https://pith.science/pith/GJYE2Z74N4TD56ICL65XVTAYU5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GJYE2Z74N4TD56ICL65XVTAYU5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GJYE2Z74N4TD56ICL65XVTAYU5","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":"13d371846b5301c56c4e6ec8e29601e9978eeac28bbe01a1b9a22b64762fa76e","cross_cats_sorted":["cs.CY"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T15:49:40Z","title_canon_sha256":"65d819b639cfe0c88bab6c08fb06f53296c6a8b892af95deda89f64f00b5c56e"},"schema_version":"1.0","source":{"id":"2401.16247","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.16247","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2401.16247v1","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.16247","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"GJYE2Z74N4TD","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_16","alias_value":"GJYE2Z74N4TD56IC","created_at":"2026-07-05T07:38:46Z"},{"alias_kind":"pith_short_8","alias_value":"GJYE2Z74","created_at":"2026-07-05T07:38:46Z"}],"graph_snapshots":[{"event_id":"sha256:3587c9e56af949d8232400fa8256006235ed4f5a68262de2470f51589d582b45","target":"graph","created_at":"2026-07-05T07:38:46Z","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/2401.16247/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Assessing performance in Natural Language Processing is becoming increasingly complex. One particular challenge is the potential for evaluation datasets to overlap with training data, either directly or indirectly, which can lead to skewed results and overestimation of model performance. As a consequence, human evaluation is gaining increasing interest as a means to assess the performance and reliability of models. One such method is the red teaming approach, which aims to generate edge cases where a model will produce critical errors. While this methodology is becoming standard practice for g","authors_text":"Christophe Ropers, Christophe Touret, Corinne Wong, Cristian Canton Ferrer, David Dale, Gabriel Mejia Gonzalez, Ivan Evtimov, Kristina Pereyra, Marta R. Costa-juss\\`a, Pierre Andrews, Prangthip Hansanti, Seohyun Sonia Kim","cross_cats":["cs.CY"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T15:49:40Z","title":"Towards Red Teaming in Multimodal and Multilingual Translation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.16247","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:ecd508fc86cb694cde16dda343580f6f425da9bda30ca28f60db66df2f9e4209","target":"record","created_at":"2026-07-05T07:38:46Z","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":"13d371846b5301c56c4e6ec8e29601e9978eeac28bbe01a1b9a22b64762fa76e","cross_cats_sorted":["cs.CY"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-29T15:49:40Z","title_canon_sha256":"65d819b639cfe0c88bab6c08fb06f53296c6a8b892af95deda89f64f00b5c56e"},"schema_version":"1.0","source":{"id":"2401.16247","kind":"arxiv","version":1}},"canonical_sha256":"32704d67fc6f263ef9025fbb7acc18a77f922fcf7a62f288f7545aed52627d5f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32704d67fc6f263ef9025fbb7acc18a77f922fcf7a62f288f7545aed52627d5f","first_computed_at":"2026-07-05T07:38:46.597096Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:38:46.597096Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"odffuCElOqdylB/UZNymJhfhKkSJRRuDcmmpxxHEDPrWbqhwZ5vIUNQYKjN5c+HQydgriT+NMsZdv6HGvYcKBw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:38:46.597540Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.16247","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecd508fc86cb694cde16dda343580f6f425da9bda30ca28f60db66df2f9e4209","sha256:3587c9e56af949d8232400fa8256006235ed4f5a68262de2470f51589d582b45"],"state_sha256":"deaa1c772bbe60892164d9c33d39c24eccc9486efaf8d84d6ca5441847a13dd6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FEM4mZg/iVeYoKDsDsBVX4s5EBoWyknUEyV3M9hTLwqTzl8cf/K/yMhXPJgbfRib9+SCs08X/nrtP9dtNf4RCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T20:09:09.276231Z","bundle_sha256":"4d05be1130a001acc18921c74a9583bcaee151042f87a3f3f9e14d72587a15e3"}}