{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WOF55X4YPAMJO7IFMRCF36ORWR","short_pith_number":"pith:WOF55X4Y","schema_version":"1.0","canonical_sha256":"b38bdedf987818977d0564445df9d1b44f5872eb39804964456b0c4170fbf5da","source":{"kind":"arxiv","id":"2304.04052","version":1},"attestation_state":"computed","paper":{"title":"Decoder-Only or Encoder-Decoder? Interpreting Language Model as a Regularized Encoder-Decoder","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Anthony Man-Cho So, Nigel Collier, Qian Yu, Shengding Hu, Wai Lam, Zhiyuan Liu, Zihao Fu","submitted_at":"2023-04-08T15:44:29Z","abstract_excerpt":"The sequence-to-sequence (seq2seq) task aims at generating the target sequence based on the given input source sequence. Traditionally, most of the seq2seq task is resolved by the Encoder-Decoder framework which requires an encoder to encode the source sequence and a decoder to generate the target text. Recently, a bunch of new approaches have emerged that apply decoder-only language models directly to the seq2seq task. Despite the significant advancements in applying language models to the seq2seq task, there is still a lack of thorough analysis on the effectiveness of the decoder-only langua"},"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":"2304.04052","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-08T15:44:29Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"dd293bf5c97163b3afcc9974041612e166830ac3c0d3ce3e78604e5a23985a35","abstract_canon_sha256":"426a5b29018c2fd57634b832074ce15ede040a31cef74b83aaa569ec8fc7af2a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:59:10.693644Z","signature_b64":"r6xrQpw82arDVk9Jb8hO0o+2YTIB0RkQdFsWYYIkOfE7vdUetuBGAsVRA29lRM2MstsMNFIyubMhj0GYOOHkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b38bdedf987818977d0564445df9d1b44f5872eb39804964456b0c4170fbf5da","last_reissued_at":"2026-07-05T05:59:10.693211Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:59:10.693211Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Decoder-Only or Encoder-Decoder? Interpreting Language Model as a Regularized Encoder-Decoder","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Anthony Man-Cho So, Nigel Collier, Qian Yu, Shengding Hu, Wai Lam, Zhiyuan Liu, Zihao Fu","submitted_at":"2023-04-08T15:44:29Z","abstract_excerpt":"The sequence-to-sequence (seq2seq) task aims at generating the target sequence based on the given input source sequence. Traditionally, most of the seq2seq task is resolved by the Encoder-Decoder framework which requires an encoder to encode the source sequence and a decoder to generate the target text. Recently, a bunch of new approaches have emerged that apply decoder-only language models directly to the seq2seq task. Despite the significant advancements in applying language models to the seq2seq task, there is still a lack of thorough analysis on the effectiveness of the decoder-only langua"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.04052","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/2304.04052/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":"2304.04052","created_at":"2026-07-05T05:59:10.693271+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.04052v1","created_at":"2026-07-05T05:59:10.693271+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.04052","created_at":"2026-07-05T05:59:10.693271+00:00"},{"alias_kind":"pith_short_12","alias_value":"WOF55X4YPAMJ","created_at":"2026-07-05T05:59:10.693271+00:00"},{"alias_kind":"pith_short_16","alias_value":"WOF55X4YPAMJO7IF","created_at":"2026-07-05T05:59:10.693271+00:00"},{"alias_kind":"pith_short_8","alias_value":"WOF55X4Y","created_at":"2026-07-05T05:59:10.693271+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.13318","citing_title":"HapticCap: A Multimodal Dataset and Task for Understanding User Experience of Vibration Haptic Signals","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR","json":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR.json","graph_json":"https://pith.science/api/pith-number/WOF55X4YPAMJO7IFMRCF36ORWR/graph.json","events_json":"https://pith.science/api/pith-number/WOF55X4YPAMJO7IFMRCF36ORWR/events.json","paper":"https://pith.science/paper/WOF55X4Y"},"agent_actions":{"view_html":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR","download_json":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR.json","view_paper":"https://pith.science/paper/WOF55X4Y","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.04052&json=true","fetch_graph":"https://pith.science/api/pith-number/WOF55X4YPAMJO7IFMRCF36ORWR/graph.json","fetch_events":"https://pith.science/api/pith-number/WOF55X4YPAMJO7IFMRCF36ORWR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR/action/storage_attestation","attest_author":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR/action/author_attestation","sign_citation":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR/action/citation_signature","submit_replication":"https://pith.science/pith/WOF55X4YPAMJO7IFMRCF36ORWR/action/replication_record"}},"created_at":"2026-07-05T05:59:10.693271+00:00","updated_at":"2026-07-05T05:59:10.693271+00:00"}