{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:O5WX4HRAIFVYTCDQWROJ76V2BR","short_pith_number":"pith:O5WX4HRA","canonical_record":{"source":{"id":"2503.05931","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T20:57:43Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"42018a50fe6b2909f7193f2b2f99f79ede780511ed4366771a12538479cdcf55","abstract_canon_sha256":"bd866690e05f68d262eae8a92e6e14299eb391deb00bc43a3ba319118071c891"},"schema_version":"1.0"},"canonical_sha256":"776d7e1e20416b898870b45c9ffaba0c4edc455167187b6ae82928c3397bc275","source":{"kind":"arxiv","id":"2503.05931","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.05931","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"arxiv_version","alias_value":"2503.05931v2","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.05931","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_12","alias_value":"O5WX4HRAIFVY","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_16","alias_value":"O5WX4HRAIFVYTCDQ","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_8","alias_value":"O5WX4HRA","created_at":"2026-07-05T10:35:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:O5WX4HRAIFVYTCDQWROJ76V2BR","target":"record","payload":{"canonical_record":{"source":{"id":"2503.05931","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T20:57:43Z","cross_cats_sorted":["eess.AS"],"title_canon_sha256":"42018a50fe6b2909f7193f2b2f99f79ede780511ed4366771a12538479cdcf55","abstract_canon_sha256":"bd866690e05f68d262eae8a92e6e14299eb391deb00bc43a3ba319118071c891"},"schema_version":"1.0"},"canonical_sha256":"776d7e1e20416b898870b45c9ffaba0c4edc455167187b6ae82928c3397bc275","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:35:47.489044Z","signature_b64":"+3W604T5RiEFey8zLktB/vE/Sds8yIZ9QXjFJz4Bgo982ciFqwAGfOXoSS/5TpT5ZuzegdtuB4AQRtgtpxs5Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"776d7e1e20416b898870b45c9ffaba0c4edc455167187b6ae82928c3397bc275","last_reissued_at":"2026-07-05T10:35:47.488555Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:35:47.488555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.05931","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:35:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aHb93YiF2qv6yJMLU/wjRYHkOaCjzjNETy37e9KQ1aKeu5bp7eaRAnywzrgeX1W+gXeFAQpSuCBQMu7girOICw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T07:21:00.956352Z"},"content_sha256":"1c3baa0d9c8ad3aff471c2d19dd1eb721c5bc61be0df51b32661ceafd2fd8a1f","schema_version":"1.0","event_id":"sha256:1c3baa0d9c8ad3aff471c2d19dd1eb721c5bc61be0df51b32661ceafd2fd8a1f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:O5WX4HRAIFVYTCDQWROJ76V2BR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training and Inference Efficiency of Encoder-Decoder Speech Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.AS"],"primary_cat":"cs.CL","authors_text":"Ankita Pasad, Boris Ginsburg, Daniel Galvez, Jagadeesh Balam, Ke Hu, Krishna C. Puvvada, Kunal Dhawan, Nithin Rao Koluguri, Piotr \\.Zelasko, Vitaly Lavrukhin","submitted_at":"2025-03-07T20:57:43Z","abstract_excerpt":"Attention encoder-decoder model architecture is the backbone of several recent top performing foundation speech models: Whisper, Seamless, OWSM, and Canary-1B. However, the reported data and compute requirements for their training are prohibitive for many in the research community. In this work, we focus on the efficiency angle and ask the questions of whether we are training these speech models efficiently, and what can we do to improve? We argue that a major, if not the most severe, detrimental factor for training efficiency is related to the sampling strategy of sequential data. We show tha"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.05931","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/2503.05931/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":"2026-07-05T10:35:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"d3SW/sAHjRMxFjyswLkpDxfMFSlS91rA4Ml20sfOhWiAEgqxH4f3Caw80FXNyV4oB4jt9YBEFliiDKgGlX5xAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T07:21:00.957001Z"},"content_sha256":"88603d4055e777ad960ad6a90b4a3b87e6445071e5eeb780a5d43634759503c7","schema_version":"1.0","event_id":"sha256:88603d4055e777ad960ad6a90b4a3b87e6445071e5eeb780a5d43634759503c7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/bundle.json","state_url":"https://pith.science/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/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":"2026-07-07T07:21:00Z","links":{"resolver":"https://pith.science/pith/O5WX4HRAIFVYTCDQWROJ76V2BR","bundle":"https://pith.science/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/bundle.json","state":"https://pith.science/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/O5WX4HRAIFVYTCDQWROJ76V2BR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:O5WX4HRAIFVYTCDQWROJ76V2BR","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":"bd866690e05f68d262eae8a92e6e14299eb391deb00bc43a3ba319118071c891","cross_cats_sorted":["eess.AS"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T20:57:43Z","title_canon_sha256":"42018a50fe6b2909f7193f2b2f99f79ede780511ed4366771a12538479cdcf55"},"schema_version":"1.0","source":{"id":"2503.05931","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.05931","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"arxiv_version","alias_value":"2503.05931v2","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.05931","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_12","alias_value":"O5WX4HRAIFVY","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_16","alias_value":"O5WX4HRAIFVYTCDQ","created_at":"2026-07-05T10:35:47Z"},{"alias_kind":"pith_short_8","alias_value":"O5WX4HRA","created_at":"2026-07-05T10:35:47Z"}],"graph_snapshots":[{"event_id":"sha256:88603d4055e777ad960ad6a90b4a3b87e6445071e5eeb780a5d43634759503c7","target":"graph","created_at":"2026-07-05T10:35:47Z","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/2503.05931/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Attention encoder-decoder model architecture is the backbone of several recent top performing foundation speech models: Whisper, Seamless, OWSM, and Canary-1B. However, the reported data and compute requirements for their training are prohibitive for many in the research community. In this work, we focus on the efficiency angle and ask the questions of whether we are training these speech models efficiently, and what can we do to improve? We argue that a major, if not the most severe, detrimental factor for training efficiency is related to the sampling strategy of sequential data. We show tha","authors_text":"Ankita Pasad, Boris Ginsburg, Daniel Galvez, Jagadeesh Balam, Ke Hu, Krishna C. Puvvada, Kunal Dhawan, Nithin Rao Koluguri, Piotr \\.Zelasko, Vitaly Lavrukhin","cross_cats":["eess.AS"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T20:57:43Z","title":"Training and Inference Efficiency of Encoder-Decoder Speech Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.05931","kind":"arxiv","version":2},"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":"sha256:1c3baa0d9c8ad3aff471c2d19dd1eb721c5bc61be0df51b32661ceafd2fd8a1f","target":"record","created_at":"2026-07-05T10:35:47Z","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":"bd866690e05f68d262eae8a92e6e14299eb391deb00bc43a3ba319118071c891","cross_cats_sorted":["eess.AS"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T20:57:43Z","title_canon_sha256":"42018a50fe6b2909f7193f2b2f99f79ede780511ed4366771a12538479cdcf55"},"schema_version":"1.0","source":{"id":"2503.05931","kind":"arxiv","version":2}},"canonical_sha256":"776d7e1e20416b898870b45c9ffaba0c4edc455167187b6ae82928c3397bc275","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"776d7e1e20416b898870b45c9ffaba0c4edc455167187b6ae82928c3397bc275","first_computed_at":"2026-07-05T10:35:47.488555Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:35:47.488555Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+3W604T5RiEFey8zLktB/vE/Sds8yIZ9QXjFJz4Bgo982ciFqwAGfOXoSS/5TpT5ZuzegdtuB4AQRtgtpxs5Bw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:35:47.489044Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.05931","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1c3baa0d9c8ad3aff471c2d19dd1eb721c5bc61be0df51b32661ceafd2fd8a1f","sha256:88603d4055e777ad960ad6a90b4a3b87e6445071e5eeb780a5d43634759503c7"],"state_sha256":"685cb583566af1fca9dc7ca9b1a61abd5683f53d6f5cc4edbbf096ecc3f07605"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Wv8GJ7eFwiaxPIuSAuMOg29vjFcer38skQEvlVtw5fVSb5O6uDrRth5XZl4aK1/Ed5LIg3+ZW1tk4PoO1KxjCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T07:21:00.961379Z","bundle_sha256":"7f0520dfe03a2d34b567aa6cb1db7386c68de8759bc2665c1eee9d93b72b5e91"}}