{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:LKHMTHKZBMYWWHBMVDLCYSLLX3","short_pith_number":"pith:LKHMTHKZ","schema_version":"1.0","canonical_sha256":"5a8ec99d590b316b1c2ca8d62c496bbefae527a2ac577798ec4fb4824717f932","source":{"kind":"arxiv","id":"2004.02984","version":2},"attestation_state":"computed","paper":{"title":"MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Denny Zhou, Hongkun Yu, Renjie Liu, Xiaodan Song, Yiming Yang, Zhiqing Sun","submitted_at":"2020-04-06T20:20:58Z","abstract_excerpt":"Natural Language Processing (NLP) has recently achieved great success by using huge pre-trained models with hundreds of millions of parameters. However, these models suffer from heavy model sizes and high latency such that they cannot be deployed to resource-limited mobile devices. In this paper, we propose MobileBERT for compressing and accelerating the popular BERT model. Like the original BERT, MobileBERT is task-agnostic, that is, it can be generically applied to various downstream NLP tasks via simple fine-tuning. Basically, MobileBERT is a thin version of BERT_LARGE, while equipped with "},"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":"2004.02984","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-04-06T20:20:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ed9156c6a1426059bcf1a0c57fcad039f209fdf2a1d9524125c27ca475c163fb","abstract_canon_sha256":"367fafd94abd634da33b3477e8efdbfaa62dd500caa8634418e8bce5ea545b59"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:55:20.912358Z","signature_b64":"enXh8IRrCvZS6rGJ82lJPi/NxV94Fy7MCIcV3LvJWEwH2Zay0bYP/gPImWoOzp++RtqB0hKSbIMvVDgN5kP+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a8ec99d590b316b1c2ca8d62c496bbefae527a2ac577798ec4fb4824717f932","last_reissued_at":"2026-07-05T00:55:20.911920Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:55:20.911920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MobileBERT: a Compact Task-Agnostic BERT for Resource-Limited Devices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Denny Zhou, Hongkun Yu, Renjie Liu, Xiaodan Song, Yiming Yang, Zhiqing Sun","submitted_at":"2020-04-06T20:20:58Z","abstract_excerpt":"Natural Language Processing (NLP) has recently achieved great success by using huge pre-trained models with hundreds of millions of parameters. However, these models suffer from heavy model sizes and high latency such that they cannot be deployed to resource-limited mobile devices. In this paper, we propose MobileBERT for compressing and accelerating the popular BERT model. Like the original BERT, MobileBERT is task-agnostic, that is, it can be generically applied to various downstream NLP tasks via simple fine-tuning. Basically, MobileBERT is a thin version of BERT_LARGE, while equipped with "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.02984","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/2004.02984/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":"2004.02984","created_at":"2026-07-05T00:55:20.911975+00:00"},{"alias_kind":"arxiv_version","alias_value":"2004.02984v2","created_at":"2026-07-05T00:55:20.911975+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.02984","created_at":"2026-07-05T00:55:20.911975+00:00"},{"alias_kind":"pith_short_12","alias_value":"LKHMTHKZBMYW","created_at":"2026-07-05T00:55:20.911975+00:00"},{"alias_kind":"pith_short_16","alias_value":"LKHMTHKZBMYWWHBM","created_at":"2026-07-05T00:55:20.911975+00:00"},{"alias_kind":"pith_short_8","alias_value":"LKHMTHKZ","created_at":"2026-07-05T00:55:20.911975+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02836","citing_title":"Fast Transformer Inference on ARM-Based HMPSoCs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30062","citing_title":"Little Brains, Big Feats: Exploring Compact Language Models","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2305.07759","citing_title":"TinyStories: How Small Can Language Models Be and Still Speak Coherent English?","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2505.06907","citing_title":"A Survey on Foundation Models for Personalized Federated Intelligence","ref_index":249,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20551","citing_title":"Faster or Stronger: Towards Flexible Visual Place Recognition via Weighted Aggregation and Token Pruning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2509.12539","citing_title":"LEAF: Knowledge Distillation of Text Embedding Models with Teacher-Aligned Representations","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2512.10998","citing_title":"SCOUT: A Defense Against Data Poisoning Attacks in Fine-Tuned Language Models","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18452","citing_title":"ESsEN: Training Compact Discriminative Vision-Language Transformers in a Low-Resource Setting","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17286","citing_title":"Depth Adaptive Efficient Visual Autoregressive Modeling","ref_index":53,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3","json":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3.json","graph_json":"https://pith.science/api/pith-number/LKHMTHKZBMYWWHBMVDLCYSLLX3/graph.json","events_json":"https://pith.science/api/pith-number/LKHMTHKZBMYWWHBMVDLCYSLLX3/events.json","paper":"https://pith.science/paper/LKHMTHKZ"},"agent_actions":{"view_html":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3","download_json":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3.json","view_paper":"https://pith.science/paper/LKHMTHKZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2004.02984&json=true","fetch_graph":"https://pith.science/api/pith-number/LKHMTHKZBMYWWHBMVDLCYSLLX3/graph.json","fetch_events":"https://pith.science/api/pith-number/LKHMTHKZBMYWWHBMVDLCYSLLX3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3/action/storage_attestation","attest_author":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3/action/author_attestation","sign_citation":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3/action/citation_signature","submit_replication":"https://pith.science/pith/LKHMTHKZBMYWWHBMVDLCYSLLX3/action/replication_record"}},"created_at":"2026-07-05T00:55:20.911975+00:00","updated_at":"2026-07-05T00:55:20.911975+00:00"}