{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:MKPXBF2DNN2LTNOXSJIA37GUG7","short_pith_number":"pith:MKPXBF2D","schema_version":"1.0","canonical_sha256":"629f7097436b74b9b5d792500dfcd437db30832add21658f976ca98709a65d21","source":{"kind":"arxiv","id":"2306.07650","version":1},"attestation_state":"computed","paper":{"title":"Modality Adaption or Regularization? A Case Study on End-to-End Speech Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chen Xu, Jingbo Zhu, Tong Xiao, Yuchen Han","submitted_at":"2023-06-13T09:42:48Z","abstract_excerpt":"Pre-training and fine-tuning is a paradigm for alleviating the data scarcity problem in end-to-end speech translation (E2E ST). The commonplace \"modality gap\" between speech and text data often leads to inconsistent inputs between pre-training and fine-tuning. However, we observe that this gap occurs in the early stages of fine-tuning, but does not have a major impact on the final performance. On the other hand, we find that there has another gap, which we call the \"capacity gap\": high resource tasks (such as ASR and MT) always require a large model to fit, when the model is reused for a low r"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.07650","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-13T09:42:48Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"be3dfe68b5896b718e11f48965a31b1145b7e3a82d34325f1a430f1abf6ba74f","abstract_canon_sha256":"e18a5dbadb3b99ebf32c1b47d3a94fcca8dea937b705d107f5b6a12c2ff12f2a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:20:07.241028Z","signature_b64":"xjXYl6hTNv3bkyjxlERyzitHCr9iH3x5JXx+m21I/V1rzEntlk5pzqneUoI15FPqSdGCERA+d0qD2LXfXsEbAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"629f7097436b74b9b5d792500dfcd437db30832add21658f976ca98709a65d21","last_reissued_at":"2026-07-05T06:20:07.240697Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:20:07.240697Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Modality Adaption or Regularization? A Case Study on End-to-End Speech Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Chen Xu, Jingbo Zhu, Tong Xiao, Yuchen Han","submitted_at":"2023-06-13T09:42:48Z","abstract_excerpt":"Pre-training and fine-tuning is a paradigm for alleviating the data scarcity problem in end-to-end speech translation (E2E ST). The commonplace \"modality gap\" between speech and text data often leads to inconsistent inputs between pre-training and fine-tuning. However, we observe that this gap occurs in the early stages of fine-tuning, but does not have a major impact on the final performance. On the other hand, we find that there has another gap, which we call the \"capacity gap\": high resource tasks (such as ASR and MT) always require a large model to fit, when the model is reused for a low r"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.07650","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/2306.07650/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.07650","created_at":"2026-07-05T06:20:07.240756+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.07650v1","created_at":"2026-07-05T06:20:07.240756+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.07650","created_at":"2026-07-05T06:20:07.240756+00:00"},{"alias_kind":"pith_short_12","alias_value":"MKPXBF2DNN2L","created_at":"2026-07-05T06:20:07.240756+00:00"},{"alias_kind":"pith_short_16","alias_value":"MKPXBF2DNN2LTNOX","created_at":"2026-07-05T06:20:07.240756+00:00"},{"alias_kind":"pith_short_8","alias_value":"MKPXBF2D","created_at":"2026-07-05T06:20:07.240756+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7","json":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7.json","graph_json":"https://pith.science/api/pith-number/MKPXBF2DNN2LTNOXSJIA37GUG7/graph.json","events_json":"https://pith.science/api/pith-number/MKPXBF2DNN2LTNOXSJIA37GUG7/events.json","paper":"https://pith.science/paper/MKPXBF2D"},"agent_actions":{"view_html":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7","download_json":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7.json","view_paper":"https://pith.science/paper/MKPXBF2D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.07650&json=true","fetch_graph":"https://pith.science/api/pith-number/MKPXBF2DNN2LTNOXSJIA37GUG7/graph.json","fetch_events":"https://pith.science/api/pith-number/MKPXBF2DNN2LTNOXSJIA37GUG7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7/action/storage_attestation","attest_author":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7/action/author_attestation","sign_citation":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7/action/citation_signature","submit_replication":"https://pith.science/pith/MKPXBF2DNN2LTNOXSJIA37GUG7/action/replication_record"}},"created_at":"2026-07-05T06:20:07.240756+00:00","updated_at":"2026-07-05T06:20:07.240756+00:00"}