{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:AJTC57P2BDP567U6EWGKQIBS3W","short_pith_number":"pith:AJTC57P2","schema_version":"1.0","canonical_sha256":"02662efdfa08dfdf7e9e258ca82032dda5c2c271b7e24ec6765828eb44765dae","source":{"kind":"arxiv","id":"2304.00173","version":1},"attestation_state":"computed","paper":{"title":"Lego-Features: Exporting modular encoder features for streaming and deliberation ASR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ciprian Chelba, Fran\\c{c}oise Beaufays, Johan Schalkwyk, Rami Botros, Rohit Prabhavalkar, Tara N. Sainath","submitted_at":"2023-03-31T23:33:21Z","abstract_excerpt":"In end-to-end (E2E) speech recognition models, a representational tight-coupling inevitably emerges between the encoder and the decoder. We build upon recent work that has begun to explore building encoders with modular encoded representations, such that encoders and decoders from different models can be stitched together in a zero-shot manner without further fine-tuning. While previous research only addresses full-context speech models, we explore the problem in a streaming setting as well. Our framework builds on top of existing encoded representations, converting them to modular features, d"},"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":"2304.00173","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-03-31T23:33:21Z","cross_cats_sorted":["cs.AI","cs.SD","eess.AS"],"title_canon_sha256":"eb68cf7d87e1b5cfcdc5684a209acc71d2517504caca3a162abc5ff4e2f781a1","abstract_canon_sha256":"3ba138e6a41438421c55d709fe920a1a8f71bbf2b6feda695848b5583e96edd1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:56:59.547848Z","signature_b64":"gQUheIJvLH0i3AwZgoRWmd+K8nUTFDXw0KlLN7XB9kWa9pVFpIfr5qX+KyLWh+1xjVwLY3yGijildQe1HTn7BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"02662efdfa08dfdf7e9e258ca82032dda5c2c271b7e24ec6765828eb44765dae","last_reissued_at":"2026-07-05T05:56:59.547446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:56:59.547446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lego-Features: Exporting modular encoder features for streaming and deliberation ASR","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Ciprian Chelba, Fran\\c{c}oise Beaufays, Johan Schalkwyk, Rami Botros, Rohit Prabhavalkar, Tara N. Sainath","submitted_at":"2023-03-31T23:33:21Z","abstract_excerpt":"In end-to-end (E2E) speech recognition models, a representational tight-coupling inevitably emerges between the encoder and the decoder. We build upon recent work that has begun to explore building encoders with modular encoded representations, such that encoders and decoders from different models can be stitched together in a zero-shot manner without further fine-tuning. While previous research only addresses full-context speech models, we explore the problem in a streaming setting as well. Our framework builds on top of existing encoded representations, converting them to modular features, d"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.00173","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/2304.00173/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":"2304.00173","created_at":"2026-07-05T05:56:59.547505+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.00173v1","created_at":"2026-07-05T05:56:59.547505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.00173","created_at":"2026-07-05T05:56:59.547505+00:00"},{"alias_kind":"pith_short_12","alias_value":"AJTC57P2BDP5","created_at":"2026-07-05T05:56:59.547505+00:00"},{"alias_kind":"pith_short_16","alias_value":"AJTC57P2BDP567U6","created_at":"2026-07-05T05:56:59.547505+00:00"},{"alias_kind":"pith_short_8","alias_value":"AJTC57P2","created_at":"2026-07-05T05:56:59.547505+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/AJTC57P2BDP567U6EWGKQIBS3W","json":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W.json","graph_json":"https://pith.science/api/pith-number/AJTC57P2BDP567U6EWGKQIBS3W/graph.json","events_json":"https://pith.science/api/pith-number/AJTC57P2BDP567U6EWGKQIBS3W/events.json","paper":"https://pith.science/paper/AJTC57P2"},"agent_actions":{"view_html":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W","download_json":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W.json","view_paper":"https://pith.science/paper/AJTC57P2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.00173&json=true","fetch_graph":"https://pith.science/api/pith-number/AJTC57P2BDP567U6EWGKQIBS3W/graph.json","fetch_events":"https://pith.science/api/pith-number/AJTC57P2BDP567U6EWGKQIBS3W/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W/action/storage_attestation","attest_author":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W/action/author_attestation","sign_citation":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W/action/citation_signature","submit_replication":"https://pith.science/pith/AJTC57P2BDP567U6EWGKQIBS3W/action/replication_record"}},"created_at":"2026-07-05T05:56:59.547505+00:00","updated_at":"2026-07-05T05:56:59.547505+00:00"}