{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:4S4RCNEYL2QA3Y7S537U3RJL3S","short_pith_number":"pith:4S4RCNEY","canonical_record":{"source":{"id":"2010.12829","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T08:15:08Z","cross_cats_sorted":[],"title_canon_sha256":"a11b6798b37529fad5649910c5b8a4ecc473568a668d6e1d9285b6c9ba81239c","abstract_canon_sha256":"9da323f3cb842085e31b56d9618515fd1af7e79a9a8fa67e2ac4e25d0357607a"},"schema_version":"1.0"},"canonical_sha256":"e4b91134985ea00de3f2eeff4dc52bdcadb267d85d5f9023dad591569fe83db8","source":{"kind":"arxiv","id":"2010.12829","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.12829","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2010.12829v4","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.12829","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"4S4RCNEYL2QA","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"4S4RCNEYL2QA3Y7S","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"4S4RCNEY","created_at":"2026-07-05T02:04:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:4S4RCNEYL2QA3Y7S537U3RJL3S","target":"record","payload":{"canonical_record":{"source":{"id":"2010.12829","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T08:15:08Z","cross_cats_sorted":[],"title_canon_sha256":"a11b6798b37529fad5649910c5b8a4ecc473568a668d6e1d9285b6c9ba81239c","abstract_canon_sha256":"9da323f3cb842085e31b56d9618515fd1af7e79a9a8fa67e2ac4e25d0357607a"},"schema_version":"1.0"},"canonical_sha256":"e4b91134985ea00de3f2eeff4dc52bdcadb267d85d5f9023dad591569fe83db8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:04:12.433229Z","signature_b64":"OSZWF0wX6ky+pmxQ8beKoYjw4XKZw/CIPL8Ou4bA51NiuoghNVLcbYd8m2Pg1GoMQgnzPpsRUqNws0LQ9anSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4b91134985ea00de3f2eeff4dc52bdcadb267d85d5f9023dad591569fe83db8","last_reissued_at":"2026-07-05T02:04:12.432738Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:04:12.432738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.12829","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ODYqMOHlhUk6JUvEJkFHO6Y//M9LsvP6dLhKjcenbHk+YmAnymoEC4CPFfsa2ES5zA9I9V1O/WXPuIYILjTeCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T01:20:50.715912Z"},"content_sha256":"0705ff882c43a1848781b9199d7a3ceb6bb5c46b0900e42bbd89b752341d9607","schema_version":"1.0","event_id":"sha256:0705ff882c43a1848781b9199d7a3ceb6bb5c46b0900e42bbd89b752341d9607"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:4S4RCNEYL2QA3Y7S537U3RJL3S","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multilingual Speech Translation with Efficient Finetuning of Pretrained Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alexei Baevski, Alexis Conneau, Changhan Wang, Chau Tran, Juan Pino, Michael Auli, Xian Li, Yun Tang, Yuqing Tang","submitted_at":"2020-10-24T08:15:08Z","abstract_excerpt":"We present a simple yet effective approach to build multilingual speech-to-text (ST) translation by efficient transfer learning from pretrained speech encoder and text decoder. Our key finding is that a minimalistic LNA (LayerNorm and Attention) finetuning can achieve zero-shot crosslingual and cross-modality transfer ability by only finetuning less than 10% of the pretrained parameters. This enables effectively leveraging large pretrained models with low training cost. Using wav2vec 2.0 for acoustic modeling, and mBART for multilingual text generation, our approach advanced the new state-of-t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.12829","kind":"arxiv","version":4},"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/2010.12829/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":"2026-07-05T02:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"742QkWD8vvrm6t2dmd29QhVZZYeehkkSU97nHnRWTszc9v0VWNgAqpi8dem2+fL1su73/bJF2t4B1VTtkLBLCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T01:20:50.716414Z"},"content_sha256":"69dc660aff20e76b2160474de49a818165b159a97111759ec36183280b98079b","schema_version":"1.0","event_id":"sha256:69dc660aff20e76b2160474de49a818165b159a97111759ec36183280b98079b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/bundle.json","state_url":"https://pith.science/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/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":"2026-08-07T01:20:50Z","links":{"resolver":"https://pith.science/pith/4S4RCNEYL2QA3Y7S537U3RJL3S","bundle":"https://pith.science/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/bundle.json","state":"https://pith.science/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4S4RCNEYL2QA3Y7S537U3RJL3S/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:4S4RCNEYL2QA3Y7S537U3RJL3S","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":"9da323f3cb842085e31b56d9618515fd1af7e79a9a8fa67e2ac4e25d0357607a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T08:15:08Z","title_canon_sha256":"a11b6798b37529fad5649910c5b8a4ecc473568a668d6e1d9285b6c9ba81239c"},"schema_version":"1.0","source":{"id":"2010.12829","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.12829","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2010.12829v4","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.12829","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"4S4RCNEYL2QA","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"4S4RCNEYL2QA3Y7S","created_at":"2026-07-05T02:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"4S4RCNEY","created_at":"2026-07-05T02:04:12Z"}],"graph_snapshots":[{"event_id":"sha256:69dc660aff20e76b2160474de49a818165b159a97111759ec36183280b98079b","target":"graph","created_at":"2026-07-05T02:04:12Z","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/2010.12829/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present a simple yet effective approach to build multilingual speech-to-text (ST) translation by efficient transfer learning from pretrained speech encoder and text decoder. Our key finding is that a minimalistic LNA (LayerNorm and Attention) finetuning can achieve zero-shot crosslingual and cross-modality transfer ability by only finetuning less than 10% of the pretrained parameters. This enables effectively leveraging large pretrained models with low training cost. Using wav2vec 2.0 for acoustic modeling, and mBART for multilingual text generation, our approach advanced the new state-of-t","authors_text":"Alexei Baevski, Alexis Conneau, Changhan Wang, Chau Tran, Juan Pino, Michael Auli, Xian Li, Yun Tang, Yuqing Tang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T08:15:08Z","title":"Multilingual Speech Translation with Efficient Finetuning of Pretrained Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.12829","kind":"arxiv","version":4},"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":"sha256:0705ff882c43a1848781b9199d7a3ceb6bb5c46b0900e42bbd89b752341d9607","target":"record","created_at":"2026-07-05T02:04:12Z","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":"9da323f3cb842085e31b56d9618515fd1af7e79a9a8fa67e2ac4e25d0357607a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-10-24T08:15:08Z","title_canon_sha256":"a11b6798b37529fad5649910c5b8a4ecc473568a668d6e1d9285b6c9ba81239c"},"schema_version":"1.0","source":{"id":"2010.12829","kind":"arxiv","version":4}},"canonical_sha256":"e4b91134985ea00de3f2eeff4dc52bdcadb267d85d5f9023dad591569fe83db8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4b91134985ea00de3f2eeff4dc52bdcadb267d85d5f9023dad591569fe83db8","first_computed_at":"2026-07-05T02:04:12.432738Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:04:12.432738Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"OSZWF0wX6ky+pmxQ8beKoYjw4XKZw/CIPL8Ou4bA51NiuoghNVLcbYd8m2Pg1GoMQgnzPpsRUqNws0LQ9anSDg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:04:12.433229Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.12829","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0705ff882c43a1848781b9199d7a3ceb6bb5c46b0900e42bbd89b752341d9607","sha256:69dc660aff20e76b2160474de49a818165b159a97111759ec36183280b98079b"],"state_sha256":"9c7c3d2ff98f17441395c5ebdfb8dc2b3a8b105afe450bac60e9ad3f110215db"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XqHaeXBThfbm/7n6Th7FXlCAIheN2+XAO5BQSbBZTzTY1Oe3sT7dwODQW3kHOwiUVQqt71dG7FkKDxVj5iYxDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T01:20:50.720301Z","bundle_sha256":"08b5515654a5b20aa982a258c5399312698ea8d905f6f0bbee64a73a8091fd68"}}