{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:5ACNDPYLSRVYQKQAGMDNLMWJB3","short_pith_number":"pith:5ACNDPYL","canonical_record":{"source":{"id":"1907.11065","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-25T14:03:06Z","cross_cats_sorted":[],"title_canon_sha256":"850599a9a9bc14d1598694a3f36bf24ba28307470c07392723f7c8553da122d3","abstract_canon_sha256":"0fe7f70b5ed4019bab8be73db71f33d4342e6b720b88cd4f0e74ccef4cb4abb2"},"schema_version":"1.0"},"canonical_sha256":"e804d1bf0b946b882a003306d5b2c90ed296c255d49eb77aebde06a9ee1d960d","source":{"kind":"arxiv","id":"1907.11065","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.11065","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"arxiv_version","alias_value":"1907.11065v2","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.11065","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"pith_short_12","alias_value":"5ACNDPYLSRVY","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"5ACNDPYLSRVYQKQA","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"5ACNDPYL","created_at":"2026-05-18T12:33:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:5ACNDPYLSRVYQKQAGMDNLMWJB3","target":"record","payload":{"canonical_record":{"source":{"id":"1907.11065","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-25T14:03:06Z","cross_cats_sorted":[],"title_canon_sha256":"850599a9a9bc14d1598694a3f36bf24ba28307470c07392723f7c8553da122d3","abstract_canon_sha256":"0fe7f70b5ed4019bab8be73db71f33d4342e6b720b88cd4f0e74ccef4cb4abb2"},"schema_version":"1.0"},"canonical_sha256":"e804d1bf0b946b882a003306d5b2c90ed296c255d49eb77aebde06a9ee1d960d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:29.781087Z","signature_b64":"CH85grI2cYk2NaYIintCgz+vOdZSVHgp8UO8HdydntpdjDvJid3uYSckSmExc+bIyiRvm9z2w2NlaTj0xvtZDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e804d1bf0b946b882a003306d5b2c90ed296c255d49eb77aebde06a9ee1d960d","last_reissued_at":"2026-05-17T23:39:29.780600Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:29.780600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.11065","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"spkV/4Krq9ot2+Qqh5kTFPAHTWBevChXPN4Ggnc1LSB85IOnxba7Bx1+mh0C/mOtUCGjNZMaBuJysoUL4d6ECQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T14:12:03.071623Z"},"content_sha256":"1f861840b3b894a9660221ed2f25a3d9fb438a6a6eac0af51d0bbc43020a476d","schema_version":"1.0","event_id":"sha256:1f861840b3b894a9660221ed2f25a3d9fb438a6a6eac0af51d0bbc43020a476d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:5ACNDPYLSRVYQKQAGMDNLMWJB3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"DropAttention: A Regularization Method for Fully-Connected Self-Attention Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Junkun Chen, Lin Zehui, Luyao Huang, Pengfei Liu, Xipeng Qiu, Xuanjing Huang","submitted_at":"2019-07-25T14:03:06Z","abstract_excerpt":"Variants dropout methods have been designed for the fully-connected layer, convolutional layer and recurrent layer in neural networks, and shown to be effective to avoid overfitting. As an appealing alternative to recurrent and convolutional layers, the fully-connected self-attention layer surprisingly lacks a specific dropout method. This paper explores the possibility of regularizing the attention weights in Transformers to prevent different contextualized feature vectors from co-adaption. Experiments on a wide range of tasks show that DropAttention can improve performance and reduce overfit"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.11065","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":""},"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":"2026-05-17T23:39:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AE9hUK8DmWv4QfOLSRjnTRpyAVgNIeG5SFdx0NzSzOwh8DZXeoAaYq4SA4oW6lLcOkaIT2kV0li+khT3pMKLBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-23T14:12:03.072338Z"},"content_sha256":"03d08f0636137ef88ddb7dd401a367c8320958bfb7ca8d41dc31cde0f7d4fad3","schema_version":"1.0","event_id":"sha256:03d08f0636137ef88ddb7dd401a367c8320958bfb7ca8d41dc31cde0f7d4fad3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/bundle.json","state_url":"https://pith.science/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/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":"2026-05-23T14:12:03Z","links":{"resolver":"https://pith.science/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3","bundle":"https://pith.science/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/bundle.json","state":"https://pith.science/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5ACNDPYLSRVYQKQAGMDNLMWJB3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:5ACNDPYLSRVYQKQAGMDNLMWJB3","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":"0fe7f70b5ed4019bab8be73db71f33d4342e6b720b88cd4f0e74ccef4cb4abb2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-25T14:03:06Z","title_canon_sha256":"850599a9a9bc14d1598694a3f36bf24ba28307470c07392723f7c8553da122d3"},"schema_version":"1.0","source":{"id":"1907.11065","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.11065","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"arxiv_version","alias_value":"1907.11065v2","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.11065","created_at":"2026-05-17T23:39:29Z"},{"alias_kind":"pith_short_12","alias_value":"5ACNDPYLSRVY","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_16","alias_value":"5ACNDPYLSRVYQKQA","created_at":"2026-05-18T12:33:10Z"},{"alias_kind":"pith_short_8","alias_value":"5ACNDPYL","created_at":"2026-05-18T12:33:10Z"}],"graph_snapshots":[{"event_id":"sha256:03d08f0636137ef88ddb7dd401a367c8320958bfb7ca8d41dc31cde0f7d4fad3","target":"graph","created_at":"2026-05-17T23:39:29Z","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":"Variants dropout methods have been designed for the fully-connected layer, convolutional layer and recurrent layer in neural networks, and shown to be effective to avoid overfitting. As an appealing alternative to recurrent and convolutional layers, the fully-connected self-attention layer surprisingly lacks a specific dropout method. This paper explores the possibility of regularizing the attention weights in Transformers to prevent different contextualized feature vectors from co-adaption. Experiments on a wide range of tasks show that DropAttention can improve performance and reduce overfit","authors_text":"Junkun Chen, Lin Zehui, Luyao Huang, Pengfei Liu, Xipeng Qiu, Xuanjing Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-25T14:03:06Z","title":"DropAttention: A Regularization Method for Fully-Connected Self-Attention Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.11065","kind":"arxiv","version":2},"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":"sha256:1f861840b3b894a9660221ed2f25a3d9fb438a6a6eac0af51d0bbc43020a476d","target":"record","created_at":"2026-05-17T23:39:29Z","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":"0fe7f70b5ed4019bab8be73db71f33d4342e6b720b88cd4f0e74ccef4cb4abb2","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-07-25T14:03:06Z","title_canon_sha256":"850599a9a9bc14d1598694a3f36bf24ba28307470c07392723f7c8553da122d3"},"schema_version":"1.0","source":{"id":"1907.11065","kind":"arxiv","version":2}},"canonical_sha256":"e804d1bf0b946b882a003306d5b2c90ed296c255d49eb77aebde06a9ee1d960d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e804d1bf0b946b882a003306d5b2c90ed296c255d49eb77aebde06a9ee1d960d","first_computed_at":"2026-05-17T23:39:29.780600Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:29.780600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"CH85grI2cYk2NaYIintCgz+vOdZSVHgp8UO8HdydntpdjDvJid3uYSckSmExc+bIyiRvm9z2w2NlaTj0xvtZDg==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:29.781087Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.11065","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1f861840b3b894a9660221ed2f25a3d9fb438a6a6eac0af51d0bbc43020a476d","sha256:03d08f0636137ef88ddb7dd401a367c8320958bfb7ca8d41dc31cde0f7d4fad3"],"state_sha256":"f21c7deb9d9a64471ffdc1a23dff3a0a16b7ab77edb4197bd4c59844dbeb0e71"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e+sk41G2YR6JTH3ubo+wktLc/wx62JL5rheZSdml7GzitdfUn35G3uCzX0KlQh6/CG0UC+MuZIo2SU524oPhDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-23T14:12:03.077093Z","bundle_sha256":"b35a15590618d46ff24069ebf527d8f36b18c39ced9a377cf21e2f195f62dfde"}}