{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:Z7UBR4VDVXTUYKCCYSZPPLAGL4","short_pith_number":"pith:Z7UBR4VD","schema_version":"1.0","canonical_sha256":"cfe818f2a3ade74c2842c4b2f7ac065f246d2dfa3bd806f9ad762047773250fd","source":{"kind":"arxiv","id":"2311.08623","version":1},"attestation_state":"computed","paper":{"title":"DEED: Dynamic Early Exit on Decoder for Accelerating Encoder-Decoder Transformer Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Pengkai Zhu, Peng Tang, R. Manmatha, Srikar Appalaraju, Tian Li, Vijay Mahadevan","submitted_at":"2023-11-15T01:01:02Z","abstract_excerpt":"Encoder-decoder transformer models have achieved great success on various vision-language (VL) tasks, but they suffer from high inference latency. Typically, the decoder takes up most of the latency because of the auto-regressive decoding. To accelerate the inference, we propose an approach of performing Dynamic Early Exit on Decoder (DEED). We build a multi-exit encoder-decoder transformer model which is trained with deep supervision so that each of its decoder layers is capable of generating plausible predictions. In addition, we leverage simple yet practical techniques, including shared gen"},"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":"2311.08623","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-15T01:01:02Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"8441c8bb715f885da3e208f6c7a99f05c16f85ec10227ac68c88c229d742388b","abstract_canon_sha256":"1458ab4ac3d3fe33b75865a7c530633dcaf0c9aff9f7c765a6209edf6574e359"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:13:00.602150Z","signature_b64":"ZE3qsJ0cyMVOJMzR7LnafYOEPOlBHneEyNL5FENAjOSPD8StS6L4l89+aerhtbHsFkUxZzqopqhgPHpNYcJmAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cfe818f2a3ade74c2842c4b2f7ac065f246d2dfa3bd806f9ad762047773250fd","last_reissued_at":"2026-07-05T07:13:00.601698Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:13:00.601698Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DEED: Dynamic Early Exit on Decoder for Accelerating Encoder-Decoder Transformer Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Pengkai Zhu, Peng Tang, R. Manmatha, Srikar Appalaraju, Tian Li, Vijay Mahadevan","submitted_at":"2023-11-15T01:01:02Z","abstract_excerpt":"Encoder-decoder transformer models have achieved great success on various vision-language (VL) tasks, but they suffer from high inference latency. Typically, the decoder takes up most of the latency because of the auto-regressive decoding. To accelerate the inference, we propose an approach of performing Dynamic Early Exit on Decoder (DEED). We build a multi-exit encoder-decoder transformer model which is trained with deep supervision so that each of its decoder layers is capable of generating plausible predictions. In addition, we leverage simple yet practical techniques, including shared gen"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08623","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/2311.08623/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":"2311.08623","created_at":"2026-07-05T07:13:00.601760+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.08623v1","created_at":"2026-07-05T07:13:00.601760+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08623","created_at":"2026-07-05T07:13:00.601760+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z7UBR4VDVXTU","created_at":"2026-07-05T07:13:00.601760+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z7UBR4VDVXTUYKCC","created_at":"2026-07-05T07:13:00.601760+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z7UBR4VD","created_at":"2026-07-05T07:13:00.601760+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/Z7UBR4VDVXTUYKCCYSZPPLAGL4","json":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4.json","graph_json":"https://pith.science/api/pith-number/Z7UBR4VDVXTUYKCCYSZPPLAGL4/graph.json","events_json":"https://pith.science/api/pith-number/Z7UBR4VDVXTUYKCCYSZPPLAGL4/events.json","paper":"https://pith.science/paper/Z7UBR4VD"},"agent_actions":{"view_html":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4","download_json":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4.json","view_paper":"https://pith.science/paper/Z7UBR4VD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.08623&json=true","fetch_graph":"https://pith.science/api/pith-number/Z7UBR4VDVXTUYKCCYSZPPLAGL4/graph.json","fetch_events":"https://pith.science/api/pith-number/Z7UBR4VDVXTUYKCCYSZPPLAGL4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4/action/storage_attestation","attest_author":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4/action/author_attestation","sign_citation":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4/action/citation_signature","submit_replication":"https://pith.science/pith/Z7UBR4VDVXTUYKCCYSZPPLAGL4/action/replication_record"}},"created_at":"2026-07-05T07:13:00.601760+00:00","updated_at":"2026-07-05T07:13:00.601760+00:00"}