{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:ZLR2YMVEMGHIGOMAK2WISDBYWR","short_pith_number":"pith:ZLR2YMVE","canonical_record":{"source":{"id":"2104.05146","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-12T00:49:51Z","cross_cats_sorted":[],"title_canon_sha256":"90a49128de9b7cafdbc98042b73265b368693ab9cad57eee62bd95242c574e76","abstract_canon_sha256":"bf2b2cb0def1cbe0537458a429e49e0765f90c201c9f6c63bfb7f3ab1ad5ac8e"},"schema_version":"1.0"},"canonical_sha256":"cae3ac32a4618e83398056ac890c38b465f278f915860797eaa356a033394a74","source":{"kind":"arxiv","id":"2104.05146","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.05146","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"arxiv_version","alias_value":"2104.05146v1","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.05146","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_12","alias_value":"ZLR2YMVEMGHI","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_16","alias_value":"ZLR2YMVEMGHIGOMA","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_8","alias_value":"ZLR2YMVE","created_at":"2026-07-05T02:31:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:ZLR2YMVEMGHIGOMAK2WISDBYWR","target":"record","payload":{"canonical_record":{"source":{"id":"2104.05146","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-12T00:49:51Z","cross_cats_sorted":[],"title_canon_sha256":"90a49128de9b7cafdbc98042b73265b368693ab9cad57eee62bd95242c574e76","abstract_canon_sha256":"bf2b2cb0def1cbe0537458a429e49e0765f90c201c9f6c63bfb7f3ab1ad5ac8e"},"schema_version":"1.0"},"canonical_sha256":"cae3ac32a4618e83398056ac890c38b465f278f915860797eaa356a033394a74","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:31:08.794900Z","signature_b64":"maDVwYy49adW1iUUmI4TY2C+2O9cwpPUHKdgrp2kZqxIMTTj66ltLopkUF5rq4MPolI0I83kHEMYu9NdmIzjBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cae3ac32a4618e83398056ac890c38b465f278f915860797eaa356a033394a74","last_reissued_at":"2026-07-05T02:31:08.794356Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:31:08.794356Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2104.05146","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":"2026-07-05T02:31:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yRnLlw25l+LdmZm4HSAL1NvCeumrFlHBd1e/cN2qPNElP2878CJ1SB9eWdCg0YVeKmqKQdk/buIBPTd2+zA1Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T06:15:27.816205Z"},"content_sha256":"575a46f9ca405fa5d44864d75037b3954b146986db880579d992427b05f73c25","schema_version":"1.0","event_id":"sha256:575a46f9ca405fa5d44864d75037b3954b146986db880579d992427b05f73c25"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:ZLR2YMVEMGHIGOMAK2WISDBYWR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Assessing Reference-Free Peer Evaluation for Machine Translation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Colin Cherry, George Foster, Markus Freitag, Sweta Agrawal","submitted_at":"2021-04-12T00:49:51Z","abstract_excerpt":"Reference-free evaluation has the potential to make machine translation evaluation substantially more scalable, allowing us to pivot easily to new languages or domains. It has been recently shown that the probabilities given by a large, multilingual model can achieve state of the art results when used as a reference-free metric. We experiment with various modifications to this model and demonstrate that by scaling it up we can match the performance of BLEU. We analyze various potential weaknesses of the approach and find that it is surprisingly robust and likely to offer reasonable performance"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.05146","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/2104.05146/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":"2026-07-05T02:31:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TxEBYjHExW5ylmZAtTt59V/Le44rGyWQc/yoOhQafLQqy8tA8HW9sooGtEuGn/cyD5esYNEKjvkO4/MUNpK4CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T06:15:27.817129Z"},"content_sha256":"0574e7a4e18b3a05601ec0d6b4a5c7fc0261d09de6c4cc80b6a8ba28c0ea4d76","schema_version":"1.0","event_id":"sha256:0574e7a4e18b3a05601ec0d6b4a5c7fc0261d09de6c4cc80b6a8ba28c0ea4d76"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/bundle.json","state_url":"https://pith.science/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/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":"2026-08-05T06:15:27Z","links":{"resolver":"https://pith.science/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR","bundle":"https://pith.science/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/bundle.json","state":"https://pith.science/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZLR2YMVEMGHIGOMAK2WISDBYWR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:ZLR2YMVEMGHIGOMAK2WISDBYWR","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":"bf2b2cb0def1cbe0537458a429e49e0765f90c201c9f6c63bfb7f3ab1ad5ac8e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-12T00:49:51Z","title_canon_sha256":"90a49128de9b7cafdbc98042b73265b368693ab9cad57eee62bd95242c574e76"},"schema_version":"1.0","source":{"id":"2104.05146","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2104.05146","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"arxiv_version","alias_value":"2104.05146v1","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.05146","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_12","alias_value":"ZLR2YMVEMGHI","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_16","alias_value":"ZLR2YMVEMGHIGOMA","created_at":"2026-07-05T02:31:08Z"},{"alias_kind":"pith_short_8","alias_value":"ZLR2YMVE","created_at":"2026-07-05T02:31:08Z"}],"graph_snapshots":[{"event_id":"sha256:0574e7a4e18b3a05601ec0d6b4a5c7fc0261d09de6c4cc80b6a8ba28c0ea4d76","target":"graph","created_at":"2026-07-05T02:31:08Z","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/2104.05146/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reference-free evaluation has the potential to make machine translation evaluation substantially more scalable, allowing us to pivot easily to new languages or domains. It has been recently shown that the probabilities given by a large, multilingual model can achieve state of the art results when used as a reference-free metric. We experiment with various modifications to this model and demonstrate that by scaling it up we can match the performance of BLEU. We analyze various potential weaknesses of the approach and find that it is surprisingly robust and likely to offer reasonable performance","authors_text":"Colin Cherry, George Foster, Markus Freitag, Sweta Agrawal","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-12T00:49:51Z","title":"Assessing Reference-Free Peer Evaluation for Machine Translation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.05146","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":"sha256:575a46f9ca405fa5d44864d75037b3954b146986db880579d992427b05f73c25","target":"record","created_at":"2026-07-05T02:31:08Z","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":"bf2b2cb0def1cbe0537458a429e49e0765f90c201c9f6c63bfb7f3ab1ad5ac8e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-12T00:49:51Z","title_canon_sha256":"90a49128de9b7cafdbc98042b73265b368693ab9cad57eee62bd95242c574e76"},"schema_version":"1.0","source":{"id":"2104.05146","kind":"arxiv","version":1}},"canonical_sha256":"cae3ac32a4618e83398056ac890c38b465f278f915860797eaa356a033394a74","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cae3ac32a4618e83398056ac890c38b465f278f915860797eaa356a033394a74","first_computed_at":"2026-07-05T02:31:08.794356Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:31:08.794356Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"maDVwYy49adW1iUUmI4TY2C+2O9cwpPUHKdgrp2kZqxIMTTj66ltLopkUF5rq4MPolI0I83kHEMYu9NdmIzjBw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:31:08.794900Z","signed_message":"canonical_sha256_bytes"},"source_id":"2104.05146","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:575a46f9ca405fa5d44864d75037b3954b146986db880579d992427b05f73c25","sha256:0574e7a4e18b3a05601ec0d6b4a5c7fc0261d09de6c4cc80b6a8ba28c0ea4d76"],"state_sha256":"8d00117f33c1d4d363caee02fd2622f44e9cb6b6a89ae2f8fefedca1f44f4c36"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cVYWKgX/91xIc81gmFBCahz1SLi823jRnfFq93noj/GImqeQOdCaUnry2OxNQbBouAKsvCsmz/B9SuLH2MNHDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T06:15:27.827598Z","bundle_sha256":"ff4bf43646807085485a121025e02fa9414fd545d9603b929a1e4370dfac9331"}}