{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:C3PLMWAFIEQNNLZXNUJTRYKET4","short_pith_number":"pith:C3PLMWAF","schema_version":"1.0","canonical_sha256":"16deb658054120d6af376d1338e1449f24cf2f615f802394f892ad020e4365b9","source":{"kind":"arxiv","id":"2502.20583","version":2},"attestation_state":"computed","paper":{"title":"LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Baris Kasikci, Jungo Kasai, Keisuke Kamahori, Noriyuki Kojima","submitted_at":"2025-02-27T22:52:21Z","abstract_excerpt":"Modern automatic speech recognition (ASR) models, such as OpenAI's Whisper, rely on deep encoder-decoder architectures, and their encoders are a critical bottleneck for efficient deployment due to high computational intensity. We introduce LiteASR, a low-rank compression scheme for ASR encoders that significantly reduces inference costs while maintaining transcription accuracy. Our approach leverages the strong low-rank properties observed in intermediate activations: by applying principal component analysis (PCA) with a small calibration dataset, we approximate linear transformations with a c"},"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":"2502.20583","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T22:52:21Z","cross_cats_sorted":["cs.AI","cs.SD","eess.AS"],"title_canon_sha256":"385a8eba24476d0055b71e7aa0e1c5caaea1db73be4858a05784c51591656fa6","abstract_canon_sha256":"263f41bbb3848af9c78dee7a80536eef97f974bca94bbe0e3dea199ea8623acb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:55.099567Z","signature_b64":"W3vQV2baMxWPIo7/XwKzodzasC8gP71fQvrG2BuZgqIW3Ri2AzyapIfmf5ukeXOfOvFVnFG6jHN6/JX4f8flBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16deb658054120d6af376d1338e1449f24cf2f615f802394f892ad020e4365b9","last_reissued_at":"2026-07-05T11:57:55.099110Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:55.099110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LiteASR: Efficient Automatic Speech Recognition with Low-Rank Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.SD","eess.AS"],"primary_cat":"cs.LG","authors_text":"Baris Kasikci, Jungo Kasai, Keisuke Kamahori, Noriyuki Kojima","submitted_at":"2025-02-27T22:52:21Z","abstract_excerpt":"Modern automatic speech recognition (ASR) models, such as OpenAI's Whisper, rely on deep encoder-decoder architectures, and their encoders are a critical bottleneck for efficient deployment due to high computational intensity. We introduce LiteASR, a low-rank compression scheme for ASR encoders that significantly reduces inference costs while maintaining transcription accuracy. Our approach leverages the strong low-rank properties observed in intermediate activations: by applying principal component analysis (PCA) with a small calibration dataset, we approximate linear transformations with a c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.20583","kind":"arxiv","version":2},"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/2502.20583/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":"2502.20583","created_at":"2026-07-05T11:57:55.099168+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.20583v2","created_at":"2026-07-05T11:57:55.099168+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.20583","created_at":"2026-07-05T11:57:55.099168+00:00"},{"alias_kind":"pith_short_12","alias_value":"C3PLMWAFIEQN","created_at":"2026-07-05T11:57:55.099168+00:00"},{"alias_kind":"pith_short_16","alias_value":"C3PLMWAFIEQNNLZX","created_at":"2026-07-05T11:57:55.099168+00:00"},{"alias_kind":"pith_short_8","alias_value":"C3PLMWAF","created_at":"2026-07-05T11:57:55.099168+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.14451","citing_title":"Adapting Whisper for Lightweight and Efficient Automatic Speech Recognition of Children for On-device Edge Applications","ref_index":35,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4","json":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4.json","graph_json":"https://pith.science/api/pith-number/C3PLMWAFIEQNNLZXNUJTRYKET4/graph.json","events_json":"https://pith.science/api/pith-number/C3PLMWAFIEQNNLZXNUJTRYKET4/events.json","paper":"https://pith.science/paper/C3PLMWAF"},"agent_actions":{"view_html":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4","download_json":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4.json","view_paper":"https://pith.science/paper/C3PLMWAF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.20583&json=true","fetch_graph":"https://pith.science/api/pith-number/C3PLMWAFIEQNNLZXNUJTRYKET4/graph.json","fetch_events":"https://pith.science/api/pith-number/C3PLMWAFIEQNNLZXNUJTRYKET4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4/action/storage_attestation","attest_author":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4/action/author_attestation","sign_citation":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4/action/citation_signature","submit_replication":"https://pith.science/pith/C3PLMWAFIEQNNLZXNUJTRYKET4/action/replication_record"}},"created_at":"2026-07-05T11:57:55.099168+00:00","updated_at":"2026-07-05T11:57:55.099168+00:00"}