{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:FBG5LJIRGMSBMMLGGN46BWADHD","short_pith_number":"pith:FBG5LJIR","canonical_record":{"source":{"id":"2301.04761","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-11T23:45:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c85bb2315e9650a6004e9cd2d97dbb3e1e964aeb2079aa82408d5bf18867e4dc","abstract_canon_sha256":"97a962a59260cb30fc88d25cebf2ad63b0aebdc267b9a79c872bc3b290a720ea"},"schema_version":"1.0"},"canonical_sha256":"284dd5a51133241631663379e0d80338d1c5d77500ad69f38475eeb4cc592fff","source":{"kind":"arxiv","id":"2301.04761","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.04761","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"arxiv_version","alias_value":"2301.04761v2","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.04761","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_12","alias_value":"FBG5LJIRGMSB","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_16","alias_value":"FBG5LJIRGMSBMMLG","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_8","alias_value":"FBG5LJIR","created_at":"2026-07-05T06:17:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:FBG5LJIRGMSBMMLGGN46BWADHD","target":"record","payload":{"canonical_record":{"source":{"id":"2301.04761","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-11T23:45:50Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"c85bb2315e9650a6004e9cd2d97dbb3e1e964aeb2079aa82408d5bf18867e4dc","abstract_canon_sha256":"97a962a59260cb30fc88d25cebf2ad63b0aebdc267b9a79c872bc3b290a720ea"},"schema_version":"1.0"},"canonical_sha256":"284dd5a51133241631663379e0d80338d1c5d77500ad69f38475eeb4cc592fff","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:17:44.607223Z","signature_b64":"0V7NdGMuyyiKGBVnZVZ2cjChT/cRhNTMMsOO9VqYxx95R1v9WkXh/fNM6qUT+8Z6hXLudTGcKc+pvx329Z9kCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"284dd5a51133241631663379e0d80338d1c5d77500ad69f38475eeb4cc592fff","last_reissued_at":"2026-07-05T06:17:44.606742Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:17:44.606742Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2301.04761","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":"2026-07-05T06:17:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gm7lkOmhLipkQXOE3cpN8gg++H167u7QFczL3nSPDZ9Hru9UTjuy/cKhaiNcrAsOhG4IwmOWO1EH1DvWaKH6Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:48:51.731964Z"},"content_sha256":"c2bea304f1050c73d2789bdbaf8c33375a5a06fb09c0e0d3f0bfcc0354bb3b0a","schema_version":"1.0","event_id":"sha256:c2bea304f1050c73d2789bdbaf8c33375a5a06fb09c0e0d3f0bfcc0354bb3b0a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:FBG5LJIRGMSBMMLGGN46BWADHD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"NarrowBERT: Accelerating Masked Language Model Pretraining and Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Daniel Cheng, Haoxin Li, Jungo Kasai, Noah A. Smith, Phillip Keung","submitted_at":"2023-01-11T23:45:50Z","abstract_excerpt":"Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become larger over time. We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than $2\\times$. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining, rather than all of the tokens as with the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.04761","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2301.04761/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:17:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xiCfR4+m48O8WOuUFDgQLuVtKLrWHiIjqaksln7wRpQP6kx2Z+kwxxrN09y0S3rLtMwCPv9LWZ86uLxUHbq7Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T05:48:51.732372Z"},"content_sha256":"1cc19b0c81ef28e63be0d21befa76e9833f65a7a45ae597ed89bf63cf265586b","schema_version":"1.0","event_id":"sha256:1cc19b0c81ef28e63be0d21befa76e9833f65a7a45ae597ed89bf63cf265586b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FBG5LJIRGMSBMMLGGN46BWADHD/bundle.json","state_url":"https://pith.science/pith/FBG5LJIRGMSBMMLGGN46BWADHD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FBG5LJIRGMSBMMLGGN46BWADHD/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":"2026-07-07T05:48:51Z","links":{"resolver":"https://pith.science/pith/FBG5LJIRGMSBMMLGGN46BWADHD","bundle":"https://pith.science/pith/FBG5LJIRGMSBMMLGGN46BWADHD/bundle.json","state":"https://pith.science/pith/FBG5LJIRGMSBMMLGGN46BWADHD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FBG5LJIRGMSBMMLGGN46BWADHD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:FBG5LJIRGMSBMMLGGN46BWADHD","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":"97a962a59260cb30fc88d25cebf2ad63b0aebdc267b9a79c872bc3b290a720ea","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-11T23:45:50Z","title_canon_sha256":"c85bb2315e9650a6004e9cd2d97dbb3e1e964aeb2079aa82408d5bf18867e4dc"},"schema_version":"1.0","source":{"id":"2301.04761","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2301.04761","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"arxiv_version","alias_value":"2301.04761v2","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.04761","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_12","alias_value":"FBG5LJIRGMSB","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_16","alias_value":"FBG5LJIRGMSBMMLG","created_at":"2026-07-05T06:17:44Z"},{"alias_kind":"pith_short_8","alias_value":"FBG5LJIR","created_at":"2026-07-05T06:17:44Z"}],"graph_snapshots":[{"event_id":"sha256:1cc19b0c81ef28e63be0d21befa76e9833f65a7a45ae597ed89bf63cf265586b","target":"graph","created_at":"2026-07-05T06:17:44Z","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/2301.04761/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-scale language model pretraining is a very successful form of self-supervised learning in natural language processing, but it is increasingly expensive to perform as the models and pretraining corpora have become larger over time. We propose NarrowBERT, a modified transformer encoder that increases the throughput for masked language model pretraining by more than $2\\times$. NarrowBERT sparsifies the transformer model such that the self-attention queries and feedforward layers only operate on the masked tokens of each sentence during pretraining, rather than all of the tokens as with the ","authors_text":"Daniel Cheng, Haoxin Li, Jungo Kasai, Noah A. Smith, Phillip Keung","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-11T23:45:50Z","title":"NarrowBERT: Accelerating Masked Language Model Pretraining and Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.04761","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:c2bea304f1050c73d2789bdbaf8c33375a5a06fb09c0e0d3f0bfcc0354bb3b0a","target":"record","created_at":"2026-07-05T06:17:44Z","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":"97a962a59260cb30fc88d25cebf2ad63b0aebdc267b9a79c872bc3b290a720ea","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-01-11T23:45:50Z","title_canon_sha256":"c85bb2315e9650a6004e9cd2d97dbb3e1e964aeb2079aa82408d5bf18867e4dc"},"schema_version":"1.0","source":{"id":"2301.04761","kind":"arxiv","version":2}},"canonical_sha256":"284dd5a51133241631663379e0d80338d1c5d77500ad69f38475eeb4cc592fff","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"284dd5a51133241631663379e0d80338d1c5d77500ad69f38475eeb4cc592fff","first_computed_at":"2026-07-05T06:17:44.606742Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:17:44.606742Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0V7NdGMuyyiKGBVnZVZ2cjChT/cRhNTMMsOO9VqYxx95R1v9WkXh/fNM6qUT+8Z6hXLudTGcKc+pvx329Z9kCw==","signature_status":"signed_v1","signed_at":"2026-07-05T06:17:44.607223Z","signed_message":"canonical_sha256_bytes"},"source_id":"2301.04761","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c2bea304f1050c73d2789bdbaf8c33375a5a06fb09c0e0d3f0bfcc0354bb3b0a","sha256:1cc19b0c81ef28e63be0d21befa76e9833f65a7a45ae597ed89bf63cf265586b"],"state_sha256":"3793843064ad8eef7e052cda66cc8aaf720c1eeb7b6406bbb275c7f4edf23334"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qEGyIc903evZMRsPnbesb6LfXGCpDV9VG/o8V01gV2n72YkU5TP1jX7QxIyjbmqyO7i1Iy7q4fWDxYZEcHllCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T05:48:51.734305Z","bundle_sha256":"cb0d77a413ee911680efbc0be9cd0b61b53ed69d78bd514fb2f98cfa114129fc"}}