{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:KI2WVOYXIUDZNZBF6X6K4JSEYQ","short_pith_number":"pith:KI2WVOYX","canonical_record":{"source":{"id":"1805.00631","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-02T05:25:43Z","cross_cats_sorted":[],"title_canon_sha256":"9df5eda7177efb9c32f58fc613af32c9e015fc521802fe90a5d6b0c18f5a4173","abstract_canon_sha256":"024c2995c137b07d96f1617f92f1b80e12394327b1135c7db49f5d9ac02556ac"},"schema_version":"1.0"},"canonical_sha256":"52356abb17450796e425f5fcae2644c43e14829190d7f8e6810c38a61195345a","source":{"kind":"arxiv","id":"1805.00631","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.00631","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"arxiv_version","alias_value":"1805.00631v3","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.00631","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"pith_short_12","alias_value":"KI2WVOYXIUDZ","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KI2WVOYXIUDZNZBF","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KI2WVOYX","created_at":"2026-05-18T12:32:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:KI2WVOYXIUDZNZBF6X6K4JSEYQ","target":"record","payload":{"canonical_record":{"source":{"id":"1805.00631","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-02T05:25:43Z","cross_cats_sorted":[],"title_canon_sha256":"9df5eda7177efb9c32f58fc613af32c9e015fc521802fe90a5d6b0c18f5a4173","abstract_canon_sha256":"024c2995c137b07d96f1617f92f1b80e12394327b1135c7db49f5d9ac02556ac"},"schema_version":"1.0"},"canonical_sha256":"52356abb17450796e425f5fcae2644c43e14829190d7f8e6810c38a61195345a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:16:41.729326Z","signature_b64":"4DLv8yETZzZ9gfB4DFqugMyEhgj0tcD0pEBpU6QgzlPkISdLihmzi1NPZVFTm2oVvER+lDeBYhYTQSpADQbUCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"52356abb17450796e425f5fcae2644c43e14829190d7f8e6810c38a61195345a","last_reissued_at":"2026-05-18T00:16:41.728828Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:16:41.728828Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1805.00631","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:16:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o288pcKwe2+4VyY7rSS5kq21POWq55xblCKFGOygeLoRXcw1hKl63jg0evU/QbIhqe+YOzBmkVTL4rqwTGPYCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T00:06:52.850248Z"},"content_sha256":"8f90cf5d7aa16e1c210aeae0d68ebb155f532fa80fbad98a447f1a0de0476561","schema_version":"1.0","event_id":"sha256:8f90cf5d7aa16e1c210aeae0d68ebb155f532fa80fbad98a447f1a0de0476561"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:KI2WVOYXIUDZNZBF6X6K4JSEYQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Accelerating Neural Transformer via an Average Attention Network","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Biao Zhang, Deyi Xiong, Jinsong Su","submitted_at":"2018-05-02T05:25:43Z","abstract_excerpt":"With parallelizable attention networks, the neural Transformer is very fast to train. However, due to the auto-regressive architecture and self-attention in the decoder, the decoding procedure becomes slow. To alleviate this issue, we propose an average attention network as an alternative to the self-attention network in the decoder of the neural Transformer. The average attention network consists of two layers, with an average layer that models dependencies on previous positions and a gating layer that is stacked over the average layer to enhance the expressiveness of the proposed attention n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.00631","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-18T00:16:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JEO9eplzGHRVeGaB2hKuK0/9oaoLwBcS7T6biEKssOX5FS0ld0TNGYJVs9AsKs6R1D1r8GzVrsrujFX8tjMlBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T00:06:52.850918Z"},"content_sha256":"c8d7d881575e2c2f8d598cc8de3811db9f57bb8df2a9a50f0a99786e6a884dbd","schema_version":"1.0","event_id":"sha256:c8d7d881575e2c2f8d598cc8de3811db9f57bb8df2a9a50f0a99786e6a884dbd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/bundle.json","state_url":"https://pith.science/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/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":"2026-05-27T00:06:52Z","links":{"resolver":"https://pith.science/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ","bundle":"https://pith.science/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/bundle.json","state":"https://pith.science/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KI2WVOYXIUDZNZBF6X6K4JSEYQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:KI2WVOYXIUDZNZBF6X6K4JSEYQ","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":"024c2995c137b07d96f1617f92f1b80e12394327b1135c7db49f5d9ac02556ac","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-02T05:25:43Z","title_canon_sha256":"9df5eda7177efb9c32f58fc613af32c9e015fc521802fe90a5d6b0c18f5a4173"},"schema_version":"1.0","source":{"id":"1805.00631","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.00631","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"arxiv_version","alias_value":"1805.00631v3","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.00631","created_at":"2026-05-18T00:16:41Z"},{"alias_kind":"pith_short_12","alias_value":"KI2WVOYXIUDZ","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_16","alias_value":"KI2WVOYXIUDZNZBF","created_at":"2026-05-18T12:32:33Z"},{"alias_kind":"pith_short_8","alias_value":"KI2WVOYX","created_at":"2026-05-18T12:32:33Z"}],"graph_snapshots":[{"event_id":"sha256:c8d7d881575e2c2f8d598cc8de3811db9f57bb8df2a9a50f0a99786e6a884dbd","target":"graph","created_at":"2026-05-18T00:16:41Z","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"},"paper":{"abstract_excerpt":"With parallelizable attention networks, the neural Transformer is very fast to train. However, due to the auto-regressive architecture and self-attention in the decoder, the decoding procedure becomes slow. To alleviate this issue, we propose an average attention network as an alternative to the self-attention network in the decoder of the neural Transformer. The average attention network consists of two layers, with an average layer that models dependencies on previous positions and a gating layer that is stacked over the average layer to enhance the expressiveness of the proposed attention n","authors_text":"Biao Zhang, Deyi Xiong, Jinsong Su","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-02T05:25:43Z","title":"Accelerating Neural Transformer via an Average Attention Network"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.00631","kind":"arxiv","version":3},"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":"sha256:8f90cf5d7aa16e1c210aeae0d68ebb155f532fa80fbad98a447f1a0de0476561","target":"record","created_at":"2026-05-18T00:16:41Z","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":"024c2995c137b07d96f1617f92f1b80e12394327b1135c7db49f5d9ac02556ac","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-02T05:25:43Z","title_canon_sha256":"9df5eda7177efb9c32f58fc613af32c9e015fc521802fe90a5d6b0c18f5a4173"},"schema_version":"1.0","source":{"id":"1805.00631","kind":"arxiv","version":3}},"canonical_sha256":"52356abb17450796e425f5fcae2644c43e14829190d7f8e6810c38a61195345a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"52356abb17450796e425f5fcae2644c43e14829190d7f8e6810c38a61195345a","first_computed_at":"2026-05-18T00:16:41.728828Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:16:41.728828Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4DLv8yETZzZ9gfB4DFqugMyEhgj0tcD0pEBpU6QgzlPkISdLihmzi1NPZVFTm2oVvER+lDeBYhYTQSpADQbUCA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:16:41.729326Z","signed_message":"canonical_sha256_bytes"},"source_id":"1805.00631","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8f90cf5d7aa16e1c210aeae0d68ebb155f532fa80fbad98a447f1a0de0476561","sha256:c8d7d881575e2c2f8d598cc8de3811db9f57bb8df2a9a50f0a99786e6a884dbd"],"state_sha256":"76b49a9f5459f2219f5f2a0bd52aa699427af12479063c1beb846e6d9e6879e0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0cLLq0Joc98h7sAV/FnclvpqXKYipmoRdBK5CSpwDUs8QkwfcrAgGbpV14qqbyBICAkW2isjHEII5HGfx31PDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T00:06:52.853838Z","bundle_sha256":"9144b8b0d916e2829b4232182581399abd8ff4bbb7048bf46ce82f759211e72e"}}