{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:B7RDMKHK6OR56J3ITYYM2QX5A6","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":"943e4b1a8651546ae19f3fa4641ab417bb3e37b4a0fa672941c0afad5ee0887a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-09T14:03:22Z","title_canon_sha256":"8153be6c094a63a26671379d9f5aa87342555c18a565635ab2fc0098a75b0717"},"schema_version":"1.0","source":{"id":"2211.04898","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.04898","created_at":"2026-07-05T05:16:27Z"},{"alias_kind":"arxiv_version","alias_value":"2211.04898v2","created_at":"2026-07-05T05:16:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.04898","created_at":"2026-07-05T05:16:27Z"},{"alias_kind":"pith_short_12","alias_value":"B7RDMKHK6OR5","created_at":"2026-07-05T05:16:27Z"},{"alias_kind":"pith_short_16","alias_value":"B7RDMKHK6OR56J3I","created_at":"2026-07-05T05:16:27Z"},{"alias_kind":"pith_short_8","alias_value":"B7RDMKHK","created_at":"2026-07-05T05:16:27Z"}],"graph_snapshots":[{"event_id":"sha256:48a4f5e12702b53d8815f8aaffb99e253602e9c6c6085c86ce6271d05c92e2b6","target":"graph","created_at":"2026-07-05T05:16:27Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2211.04898/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The pre-training of masked language models (MLMs) consumes massive computation to achieve good results on downstream NLP tasks, resulting in a large carbon footprint. In the vanilla MLM, the virtual tokens, [MASK]s, act as placeholders and gather the contextualized information from unmasked tokens to restore the corrupted information. It raises the question of whether we can append [MASK]s at a later layer, to reduce the sequence length for earlier layers and make the pre-training more efficient. We show: (1) [MASK]s can indeed be appended at a later layer, being disentangled from the word emb","authors_text":"Baohao Liao, Christof Monz, David Thulke, Hermann Ney, Sanjika Hewavitharana","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-09T14:03:22Z","title":"Mask More and Mask Later: Efficient Pre-training of Masked Language Models by Disentangling the [MASK] Token"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.04898","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:d48213cd92cbfc4aa99717f731f9eec9afecd7c5203404d46ed710d5889ef081","target":"record","created_at":"2026-07-05T05:16:27Z","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":"943e4b1a8651546ae19f3fa4641ab417bb3e37b4a0fa672941c0afad5ee0887a","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-09T14:03:22Z","title_canon_sha256":"8153be6c094a63a26671379d9f5aa87342555c18a565635ab2fc0098a75b0717"},"schema_version":"1.0","source":{"id":"2211.04898","kind":"arxiv","version":2}},"canonical_sha256":"0fe23628eaf3a3df27689e30cd42fd07884f3e15c318d0fe0668e6291a959ecb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0fe23628eaf3a3df27689e30cd42fd07884f3e15c318d0fe0668e6291a959ecb","first_computed_at":"2026-07-05T05:16:27.351426Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:16:27.351426Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lvImafcNns+/d0D6bsfU+YXBMHm+jHLvXUuNpNB1U3J19RuLfzM9vb5wGbh2TDClUjpvOa/WZIvabheVa2etAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T05:16:27.351912Z","signed_message":"canonical_sha256_bytes"},"source_id":"2211.04898","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d48213cd92cbfc4aa99717f731f9eec9afecd7c5203404d46ed710d5889ef081","sha256:48a4f5e12702b53d8815f8aaffb99e253602e9c6c6085c86ce6271d05c92e2b6"],"state_sha256":"7ec817e4338aab20427281b0e79a37509a8a76b4932cbd2017d1299c2bc81d45"}