{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:YKOWGFBHMIB7SALYX4JJ2NXXRC","short_pith_number":"pith:YKOWGFBH","schema_version":"1.0","canonical_sha256":"c29d6314276203f90178bf129d36f7889106e4d5c13cb7ef0ce1d578169e9faf","source":{"kind":"arxiv","id":"2201.05767","version":2},"attestation_state":"computed","paper":{"title":"Ensemble Transformer for Efficient and Accurate Ranking Tasks: an Application to Question Answering Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessandro Moschitti, Eric Lind, Luca Soldaini, Yoshitomo Matsubara","submitted_at":"2022-01-15T06:21:01Z","abstract_excerpt":"Large transformer models can highly improve Answer Sentence Selection (AS2) tasks, but their high computational costs prevent their use in many real-world applications. In this paper, we explore the following research question: How can we make the AS2 models more accurate without significantly increasing their model complexity? To address the question, we propose a Multiple Heads Student architecture (named CERBERUS), an efficient neural network designed to distill an ensemble of large transformers into a single smaller model. CERBERUS consists of two components: a stack of transformer layers "},"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":"2201.05767","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-01-15T06:21:01Z","cross_cats_sorted":[],"title_canon_sha256":"9e54a5a64cb50776f75ca61849e8cafa68e9af1e650ddb32d4409a22704a8c2b","abstract_canon_sha256":"9f07cfc4d15b9deba6082fb83c9c9410a29d25e5278ef2c8062e9f28d1bef55e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:55:50.496459Z","signature_b64":"v5YJUgDpH6KWnmhrilu3ObXMvMDzdfebE+RxyYHzsMEVm8hBPT5xKWAxHCECZsB+C4UejTh9TXKk1pkAqqtcCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c29d6314276203f90178bf129d36f7889106e4d5c13cb7ef0ce1d578169e9faf","last_reissued_at":"2026-07-05T09:55:50.495985Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:55:50.495985Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Ensemble Transformer for Efficient and Accurate Ranking Tasks: an Application to Question Answering Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alessandro Moschitti, Eric Lind, Luca Soldaini, Yoshitomo Matsubara","submitted_at":"2022-01-15T06:21:01Z","abstract_excerpt":"Large transformer models can highly improve Answer Sentence Selection (AS2) tasks, but their high computational costs prevent their use in many real-world applications. In this paper, we explore the following research question: How can we make the AS2 models more accurate without significantly increasing their model complexity? To address the question, we propose a Multiple Heads Student architecture (named CERBERUS), an efficient neural network designed to distill an ensemble of large transformers into a single smaller model. CERBERUS consists of two components: a stack of transformer layers "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.05767","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/2201.05767/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":"2201.05767","created_at":"2026-07-05T09:55:50.496056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.05767v2","created_at":"2026-07-05T09:55:50.496056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.05767","created_at":"2026-07-05T09:55:50.496056+00:00"},{"alias_kind":"pith_short_12","alias_value":"YKOWGFBHMIB7","created_at":"2026-07-05T09:55:50.496056+00:00"},{"alias_kind":"pith_short_16","alias_value":"YKOWGFBHMIB7SALY","created_at":"2026-07-05T09:55:50.496056+00:00"},{"alias_kind":"pith_short_8","alias_value":"YKOWGFBH","created_at":"2026-07-05T09:55:50.496056+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/YKOWGFBHMIB7SALYX4JJ2NXXRC","json":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC.json","graph_json":"https://pith.science/api/pith-number/YKOWGFBHMIB7SALYX4JJ2NXXRC/graph.json","events_json":"https://pith.science/api/pith-number/YKOWGFBHMIB7SALYX4JJ2NXXRC/events.json","paper":"https://pith.science/paper/YKOWGFBH"},"agent_actions":{"view_html":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC","download_json":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC.json","view_paper":"https://pith.science/paper/YKOWGFBH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.05767&json=true","fetch_graph":"https://pith.science/api/pith-number/YKOWGFBHMIB7SALYX4JJ2NXXRC/graph.json","fetch_events":"https://pith.science/api/pith-number/YKOWGFBHMIB7SALYX4JJ2NXXRC/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC/action/storage_attestation","attest_author":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC/action/author_attestation","sign_citation":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC/action/citation_signature","submit_replication":"https://pith.science/pith/YKOWGFBHMIB7SALYX4JJ2NXXRC/action/replication_record"}},"created_at":"2026-07-05T09:55:50.496056+00:00","updated_at":"2026-07-05T09:55:50.496056+00:00"}