{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:IBUHD7XVQXD2BEUFBZMX6ORSES","short_pith_number":"pith:IBUHD7XV","canonical_record":{"source":{"id":"2408.15366","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T19:03:11Z","cross_cats_sorted":[],"title_canon_sha256":"5906a06639537eda17f305bceba45f06db4b05f58dde2f727ab8b26b680876ba","abstract_canon_sha256":"9a72ea3ced3ee2cdf5fdbb53859b634fafc7e3a6f0e3060d7167f542aa389995"},"schema_version":"1.0"},"canonical_sha256":"406871fef585c7a092850e597f3a322495e52bc846fd47318cf775427afb7779","source":{"kind":"arxiv","id":"2408.15366","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.15366","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"arxiv_version","alias_value":"2408.15366v3","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.15366","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_12","alias_value":"IBUHD7XVQXD2","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_16","alias_value":"IBUHD7XVQXD2BEUF","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_8","alias_value":"IBUHD7XV","created_at":"2026-07-05T09:13:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:IBUHD7XVQXD2BEUFBZMX6ORSES","target":"record","payload":{"canonical_record":{"source":{"id":"2408.15366","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T19:03:11Z","cross_cats_sorted":[],"title_canon_sha256":"5906a06639537eda17f305bceba45f06db4b05f58dde2f727ab8b26b680876ba","abstract_canon_sha256":"9a72ea3ced3ee2cdf5fdbb53859b634fafc7e3a6f0e3060d7167f542aa389995"},"schema_version":"1.0"},"canonical_sha256":"406871fef585c7a092850e597f3a322495e52bc846fd47318cf775427afb7779","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:13:38.343243Z","signature_b64":"iqxOyCcTYG4dycE/5nNymJV6PZ5Se3kNlm+2rF3/Puzj+I8VOeN7wIT7cgrd/QMMSjJl9ouF51BuEZO3RTntAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"406871fef585c7a092850e597f3a322495e52bc846fd47318cf775427afb7779","last_reissued_at":"2026-07-05T09:13:38.342754Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:13:38.342754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2408.15366","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":"2026-07-05T09:13:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UVYUVzmwOF+lj3RDf2Lpknwv79nuGHJHcG2Wbhxmj1sOPLyDqcmMKJfPVtKb74SMTMINi3Z1XyadtrexRMvnCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:34:28.025021Z"},"content_sha256":"57f7c9df53fdcef2e93e4c421c7806dda8ead7c096400aedbf248882e09b95c7","schema_version":"1.0","event_id":"sha256:57f7c9df53fdcef2e93e4c421c7806dda8ead7c096400aedbf248882e09b95c7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:IBUHD7XVQXD2BEUFBZMX6ORSES","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Pitfalls and Outlooks in Using COMET","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Barry Haddow, Nikita Moghe, Pinzhen Chen, Tsz Kin Lam, Vil\\'em Zouhar","submitted_at":"2024-08-27T19:03:11Z","abstract_excerpt":"The COMET metric has blazed a trail in the machine translation community, given its strong correlation with human judgements of translation quality. Its success stems from being a modified pre-trained multilingual model finetuned for quality assessment. However, it being a machine learning model also gives rise to a new set of pitfalls that may not be widely known. We investigate these unexpected behaviours from three aspects: 1) technical: obsolete software versions and compute precision; 2) data: empty content, language mismatch, and translationese at test time as well as distribution and do"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.15366","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/2408.15366/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":"2026-07-05T09:13:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/HwSsbYgXkJZLKMX4haaC8srUAX+u7eyH3fdXHPHECD2zzhG/D54GsUaJmTbCPse8CGcfHGNjaDqcXW0fPvoAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T01:34:28.025907Z"},"content_sha256":"a452a24f268ebe4161abab211b0a8b8248b56b7aca5f335affbe2387bccee7d1","schema_version":"1.0","event_id":"sha256:a452a24f268ebe4161abab211b0a8b8248b56b7aca5f335affbe2387bccee7d1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/bundle.json","state_url":"https://pith.science/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/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":"2026-08-08T01:34:28Z","links":{"resolver":"https://pith.science/pith/IBUHD7XVQXD2BEUFBZMX6ORSES","bundle":"https://pith.science/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/bundle.json","state":"https://pith.science/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IBUHD7XVQXD2BEUFBZMX6ORSES/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:IBUHD7XVQXD2BEUFBZMX6ORSES","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":"9a72ea3ced3ee2cdf5fdbb53859b634fafc7e3a6f0e3060d7167f542aa389995","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T19:03:11Z","title_canon_sha256":"5906a06639537eda17f305bceba45f06db4b05f58dde2f727ab8b26b680876ba"},"schema_version":"1.0","source":{"id":"2408.15366","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2408.15366","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"arxiv_version","alias_value":"2408.15366v3","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.15366","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_12","alias_value":"IBUHD7XVQXD2","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_16","alias_value":"IBUHD7XVQXD2BEUF","created_at":"2026-07-05T09:13:38Z"},{"alias_kind":"pith_short_8","alias_value":"IBUHD7XV","created_at":"2026-07-05T09:13:38Z"}],"graph_snapshots":[{"event_id":"sha256:a452a24f268ebe4161abab211b0a8b8248b56b7aca5f335affbe2387bccee7d1","target":"graph","created_at":"2026-07-05T09:13:38Z","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/2408.15366/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The COMET metric has blazed a trail in the machine translation community, given its strong correlation with human judgements of translation quality. Its success stems from being a modified pre-trained multilingual model finetuned for quality assessment. However, it being a machine learning model also gives rise to a new set of pitfalls that may not be widely known. We investigate these unexpected behaviours from three aspects: 1) technical: obsolete software versions and compute precision; 2) data: empty content, language mismatch, and translationese at test time as well as distribution and do","authors_text":"Barry Haddow, Nikita Moghe, Pinzhen Chen, Tsz Kin Lam, Vil\\'em Zouhar","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T19:03:11Z","title":"Pitfalls and Outlooks in Using COMET"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.15366","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":"sha256:57f7c9df53fdcef2e93e4c421c7806dda8ead7c096400aedbf248882e09b95c7","target":"record","created_at":"2026-07-05T09:13:38Z","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":"9a72ea3ced3ee2cdf5fdbb53859b634fafc7e3a6f0e3060d7167f542aa389995","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-08-27T19:03:11Z","title_canon_sha256":"5906a06639537eda17f305bceba45f06db4b05f58dde2f727ab8b26b680876ba"},"schema_version":"1.0","source":{"id":"2408.15366","kind":"arxiv","version":3}},"canonical_sha256":"406871fef585c7a092850e597f3a322495e52bc846fd47318cf775427afb7779","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"406871fef585c7a092850e597f3a322495e52bc846fd47318cf775427afb7779","first_computed_at":"2026-07-05T09:13:38.342754Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:13:38.342754Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iqxOyCcTYG4dycE/5nNymJV6PZ5Se3kNlm+2rF3/Puzj+I8VOeN7wIT7cgrd/QMMSjJl9ouF51BuEZO3RTntAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:13:38.343243Z","signed_message":"canonical_sha256_bytes"},"source_id":"2408.15366","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:57f7c9df53fdcef2e93e4c421c7806dda8ead7c096400aedbf248882e09b95c7","sha256:a452a24f268ebe4161abab211b0a8b8248b56b7aca5f335affbe2387bccee7d1"],"state_sha256":"b49c08f88f3870946e6a0e6f7854a1923277ed5282eefbfc202d8a2d25576bce"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IrdJIKbcpGNofZZZhcrKFfR3ZhCtUfE21RZtmHVeVD+r5/oLpeag4Pv6IpX6Rf7US8L8GAKV78aIpXG5z5FEBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T01:34:28.031571Z","bundle_sha256":"5086aeec51ca0d3ea62fef1eaa4704768362bed8fbc9e9311cae3bd30fbb4894"}}