{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:XZHKI3FV2P4PORD33UPR36WRSU","short_pith_number":"pith:XZHKI3FV","schema_version":"1.0","canonical_sha256":"be4ea46cb5d3f8f7447bdd1f1dfad1951e512ac572a3c28a2c1782a0f102c67a","source":{"kind":"arxiv","id":"2106.15065","version":1},"attestation_state":"computed","paper":{"title":"Rethinking End-to-End Evaluation of Decomposable Tasks: A Case Study on Spoken Language Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Alan W Black, Alissa Ostapenko, Florian Metze, Shinji Watanabe, Siddhant Arora, Siddharth Dalmia, Vijay Viswanathan","submitted_at":"2021-06-29T02:53:59Z","abstract_excerpt":"Decomposable tasks are complex and comprise of a hierarchy of sub-tasks. Spoken intent prediction, for example, combines automatic speech recognition and natural language understanding. Existing benchmarks, however, typically hold out examples for only the surface-level sub-task. As a result, models with similar performance on these benchmarks may have unobserved performance differences on the other sub-tasks. To allow insightful comparisons between competitive end-to-end architectures, we propose a framework to construct robust test sets using coordinate ascent over sub-task specific utility "},"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":"2106.15065","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-06-29T02:53:59Z","cross_cats_sorted":["cs.SD","eess.AS"],"title_canon_sha256":"e94269043f3118c1892d5c78b10ecc5664089e5b3837757823e8f936158c3aca","abstract_canon_sha256":"3fc2d1ff3015884c3779f8f447d92afbdfb82bb086d1f00886f219a0958f1028"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:53:04.907752Z","signature_b64":"l7E3MQ3ZWihDiPalfXg8XoshgsQLxOMkh/1yHx+k7FTTug7Is9w1wpeAf47Yv95kH0DpnnQEXYBzW+3lXhjYBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be4ea46cb5d3f8f7447bdd1f1dfad1951e512ac572a3c28a2c1782a0f102c67a","last_reissued_at":"2026-07-05T02:53:04.907269Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:53:04.907269Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Rethinking End-to-End Evaluation of Decomposable Tasks: A Case Study on Spoken Language Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.SD","eess.AS"],"primary_cat":"cs.CL","authors_text":"Alan W Black, Alissa Ostapenko, Florian Metze, Shinji Watanabe, Siddhant Arora, Siddharth Dalmia, Vijay Viswanathan","submitted_at":"2021-06-29T02:53:59Z","abstract_excerpt":"Decomposable tasks are complex and comprise of a hierarchy of sub-tasks. Spoken intent prediction, for example, combines automatic speech recognition and natural language understanding. Existing benchmarks, however, typically hold out examples for only the surface-level sub-task. As a result, models with similar performance on these benchmarks may have unobserved performance differences on the other sub-tasks. To allow insightful comparisons between competitive end-to-end architectures, we propose a framework to construct robust test sets using coordinate ascent over sub-task specific utility "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.15065","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/2106.15065/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":"2106.15065","created_at":"2026-07-05T02:53:04.907323+00:00"},{"alias_kind":"arxiv_version","alias_value":"2106.15065v1","created_at":"2026-07-05T02:53:04.907323+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.15065","created_at":"2026-07-05T02:53:04.907323+00:00"},{"alias_kind":"pith_short_12","alias_value":"XZHKI3FV2P4P","created_at":"2026-07-05T02:53:04.907323+00:00"},{"alias_kind":"pith_short_16","alias_value":"XZHKI3FV2P4PORD3","created_at":"2026-07-05T02:53:04.907323+00:00"},{"alias_kind":"pith_short_8","alias_value":"XZHKI3FV","created_at":"2026-07-05T02:53:04.907323+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/XZHKI3FV2P4PORD33UPR36WRSU","json":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU.json","graph_json":"https://pith.science/api/pith-number/XZHKI3FV2P4PORD33UPR36WRSU/graph.json","events_json":"https://pith.science/api/pith-number/XZHKI3FV2P4PORD33UPR36WRSU/events.json","paper":"https://pith.science/paper/XZHKI3FV"},"agent_actions":{"view_html":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU","download_json":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU.json","view_paper":"https://pith.science/paper/XZHKI3FV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2106.15065&json=true","fetch_graph":"https://pith.science/api/pith-number/XZHKI3FV2P4PORD33UPR36WRSU/graph.json","fetch_events":"https://pith.science/api/pith-number/XZHKI3FV2P4PORD33UPR36WRSU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU/action/storage_attestation","attest_author":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU/action/author_attestation","sign_citation":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU/action/citation_signature","submit_replication":"https://pith.science/pith/XZHKI3FV2P4PORD33UPR36WRSU/action/replication_record"}},"created_at":"2026-07-05T02:53:04.907323+00:00","updated_at":"2026-07-05T02:53:04.907323+00:00"}