{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:QPSWRK6JTKYZCP35V7UWNYHBDS","short_pith_number":"pith:QPSWRK6J","canonical_record":{"source":{"id":"2006.15020","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-26T14:43:43Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"4e4c9bb8b4bc5b3af0d376155d1ac4201072be3331761a0149f54c4c90179e4b","abstract_canon_sha256":"a9c86996a7817d9d556bb023f96c5828c84be73922833578e80d2e1ae1544047"},"schema_version":"1.0"},"canonical_sha256":"83e568abc99ab1913f7dafe966e0e11cb40b315bf4be2968262bb0b4c84e2032","source":{"kind":"arxiv","id":"2006.15020","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.15020","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"arxiv_version","alias_value":"2006.15020v1","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.15020","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_12","alias_value":"QPSWRK6JTKYZ","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_16","alias_value":"QPSWRK6JTKYZCP35","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_8","alias_value":"QPSWRK6J","created_at":"2026-07-05T01:13:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:QPSWRK6JTKYZCP35V7UWNYHBDS","target":"record","payload":{"canonical_record":{"source":{"id":"2006.15020","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-26T14:43:43Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"4e4c9bb8b4bc5b3af0d376155d1ac4201072be3331761a0149f54c4c90179e4b","abstract_canon_sha256":"a9c86996a7817d9d556bb023f96c5828c84be73922833578e80d2e1ae1544047"},"schema_version":"1.0"},"canonical_sha256":"83e568abc99ab1913f7dafe966e0e11cb40b315bf4be2968262bb0b4c84e2032","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:13:48.992609Z","signature_b64":"1IRM+e9KXHG6HqYDkrLIvgWVzUFrPEJy2ABmxt5PPDbPPkC73BQFtG1k9KV0UwqzADa0am6bSjTp8liDGTdZAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83e568abc99ab1913f7dafe966e0e11cb40b315bf4be2968262bb0b4c84e2032","last_reissued_at":"2026-07-05T01:13:48.991778Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:13:48.991778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2006.15020","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":"2026-07-05T01:13:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qOKotk3MkUclcj2LeHq6X5srmifI54l+kNLT0U7nW0Ru6on7Kby1vqrYFHAGsSllEZ1hKc5zJDspm8cURpVCDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:09:12.768603Z"},"content_sha256":"b1090a8d10a4bc4d60586d6a98ff7c36647b067e27fe04863c2a0f9ff303ac41","schema_version":"1.0","event_id":"sha256:b1090a8d10a4bc4d60586d6a98ff7c36647b067e27fe04863c2a0f9ff303ac41"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:QPSWRK6JTKYZCP35V7UWNYHBDS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Pre-training via Paraphrasing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Armen Aghajanyan, Gargi Ghosh, Luke Zettlemoyer, Marjan Ghazvininejad, Mike Lewis, Sida Wang","submitted_at":"2020-06-26T14:43:43Z","abstract_excerpt":"We introduce MARGE, a pre-trained sequence-to-sequence model learned with an unsupervised multi-lingual multi-document paraphrasing objective. MARGE provides an alternative to the dominant masked language modeling paradigm, where we self-supervise the reconstruction of target text by retrieving a set of related texts (in many languages) and conditioning on them to maximize the likelihood of generating the original. We show it is possible to jointly learn to do retrieval and reconstruction, given only a random initialization. The objective noisily captures aspects of paraphrase, translation, mu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.15020","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/2006.15020/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":"2026-07-05T01:13:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PMx02SmwGLahpSf+WnxTYno4j3bEyc2wQLj/WqZm1CwcqfDDSwnEnGO4FSkp0Mm42ppFSo7vd72xE7nQAUjWBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T15:09:12.769126Z"},"content_sha256":"a08fbb8cec5a92bf791aa91200eec70b5eed7d620bc554b8d8aac1c9fe39998d","schema_version":"1.0","event_id":"sha256:a08fbb8cec5a92bf791aa91200eec70b5eed7d620bc554b8d8aac1c9fe39998d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/bundle.json","state_url":"https://pith.science/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/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":"2026-08-06T15:09:12Z","links":{"resolver":"https://pith.science/pith/QPSWRK6JTKYZCP35V7UWNYHBDS","bundle":"https://pith.science/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/bundle.json","state":"https://pith.science/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QPSWRK6JTKYZCP35V7UWNYHBDS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:QPSWRK6JTKYZCP35V7UWNYHBDS","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":"a9c86996a7817d9d556bb023f96c5828c84be73922833578e80d2e1ae1544047","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-26T14:43:43Z","title_canon_sha256":"4e4c9bb8b4bc5b3af0d376155d1ac4201072be3331761a0149f54c4c90179e4b"},"schema_version":"1.0","source":{"id":"2006.15020","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2006.15020","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"arxiv_version","alias_value":"2006.15020v1","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2006.15020","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_12","alias_value":"QPSWRK6JTKYZ","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_16","alias_value":"QPSWRK6JTKYZCP35","created_at":"2026-07-05T01:13:48Z"},{"alias_kind":"pith_short_8","alias_value":"QPSWRK6J","created_at":"2026-07-05T01:13:48Z"}],"graph_snapshots":[{"event_id":"sha256:a08fbb8cec5a92bf791aa91200eec70b5eed7d620bc554b8d8aac1c9fe39998d","target":"graph","created_at":"2026-07-05T01:13:48Z","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/2006.15020/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce MARGE, a pre-trained sequence-to-sequence model learned with an unsupervised multi-lingual multi-document paraphrasing objective. MARGE provides an alternative to the dominant masked language modeling paradigm, where we self-supervise the reconstruction of target text by retrieving a set of related texts (in many languages) and conditioning on them to maximize the likelihood of generating the original. We show it is possible to jointly learn to do retrieval and reconstruction, given only a random initialization. The objective noisily captures aspects of paraphrase, translation, mu","authors_text":"Armen Aghajanyan, Gargi Ghosh, Luke Zettlemoyer, Marjan Ghazvininejad, Mike Lewis, Sida Wang","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-26T14:43:43Z","title":"Pre-training via Paraphrasing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2006.15020","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:b1090a8d10a4bc4d60586d6a98ff7c36647b067e27fe04863c2a0f9ff303ac41","target":"record","created_at":"2026-07-05T01:13:48Z","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":"a9c86996a7817d9d556bb023f96c5828c84be73922833578e80d2e1ae1544047","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-06-26T14:43:43Z","title_canon_sha256":"4e4c9bb8b4bc5b3af0d376155d1ac4201072be3331761a0149f54c4c90179e4b"},"schema_version":"1.0","source":{"id":"2006.15020","kind":"arxiv","version":1}},"canonical_sha256":"83e568abc99ab1913f7dafe966e0e11cb40b315bf4be2968262bb0b4c84e2032","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"83e568abc99ab1913f7dafe966e0e11cb40b315bf4be2968262bb0b4c84e2032","first_computed_at":"2026-07-05T01:13:48.991778Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:13:48.991778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1IRM+e9KXHG6HqYDkrLIvgWVzUFrPEJy2ABmxt5PPDbPPkC73BQFtG1k9KV0UwqzADa0am6bSjTp8liDGTdZAw==","signature_status":"signed_v1","signed_at":"2026-07-05T01:13:48.992609Z","signed_message":"canonical_sha256_bytes"},"source_id":"2006.15020","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b1090a8d10a4bc4d60586d6a98ff7c36647b067e27fe04863c2a0f9ff303ac41","sha256:a08fbb8cec5a92bf791aa91200eec70b5eed7d620bc554b8d8aac1c9fe39998d"],"state_sha256":"4905cca897a95629bef408827d33441527102e8e17fb02f0ccbe3cbab7c7a269"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"duaXpXCRMUleyRz3xl5cD3zybhtWfWvO0IkrqgVqhlGfj4d73NHXqV8HMeKIhWfr+jm5zYkvPcmQGCqhS3UYDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T15:09:12.774735Z","bundle_sha256":"796feabcb5dea62fdf1441e9fc01644d710f1380767104aae9a86b43d07ce561"}}