{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:4OYOAVSFGG5O7C7LMVVHFNXRE3","short_pith_number":"pith:4OYOAVSF","schema_version":"1.0","canonical_sha256":"e3b0e0564531baef8beb656a72b6f126f6d53ec125aead414a7aa9aaf5024266","source":{"kind":"arxiv","id":"2010.13369","version":1},"attestation_state":"computed","paper":{"title":"Accelerating Training of Transformer-Based Language Models with Progressive Layer Dropping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Minjia Zhang, Yuxiong He","submitted_at":"2020-10-26T06:50:07Z","abstract_excerpt":"Recently, Transformer-based language models have demonstrated remarkable performance across many NLP domains. However, the unsupervised pre-training step of these models suffers from unbearable overall computational expenses. Current methods for accelerating the pre-training either rely on massive parallelism with advanced hardware or are not applicable to language modeling. In this work, we propose a method based on progressive layer dropping that speeds the training of Transformer-based language models, not at the cost of excessive hardware resources but from model architecture change and tr"},"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":"2010.13369","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-10-26T06:50:07Z","cross_cats_sorted":[],"title_canon_sha256":"30ffca25a6805203e68c8d708c382c79eb8b4867c3e768142ad3b5e1a46acf3a","abstract_canon_sha256":"e22e2a6be24fe95c3a4cc5af4729680421350f8d33df60b4619b789d6dc79c0e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:46:05.492878Z","signature_b64":"aURSng2v2BtgUgPjKQobHVmTTSmyGc1A9fOLeZf2oAMLNejKbpqU2ylRQEr/d7gI+gvDDyjaXCDucYhEqyaaBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3b0e0564531baef8beb656a72b6f126f6d53ec125aead414a7aa9aaf5024266","last_reissued_at":"2026-07-05T01:46:05.492335Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:46:05.492335Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Training of Transformer-Based Language Models with Progressive Layer Dropping","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Minjia Zhang, Yuxiong He","submitted_at":"2020-10-26T06:50:07Z","abstract_excerpt":"Recently, Transformer-based language models have demonstrated remarkable performance across many NLP domains. However, the unsupervised pre-training step of these models suffers from unbearable overall computational expenses. Current methods for accelerating the pre-training either rely on massive parallelism with advanced hardware or are not applicable to language modeling. In this work, we propose a method based on progressive layer dropping that speeds the training of Transformer-based language models, not at the cost of excessive hardware resources but from model architecture change and tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.13369","kind":"arxiv","version":1},"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/2010.13369/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":"2010.13369","created_at":"2026-07-05T01:46:05.492405+00:00"},{"alias_kind":"arxiv_version","alias_value":"2010.13369v1","created_at":"2026-07-05T01:46:05.492405+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.13369","created_at":"2026-07-05T01:46:05.492405+00:00"},{"alias_kind":"pith_short_12","alias_value":"4OYOAVSFGG5O","created_at":"2026-07-05T01:46:05.492405+00:00"},{"alias_kind":"pith_short_16","alias_value":"4OYOAVSFGG5O7C7L","created_at":"2026-07-05T01:46:05.492405+00:00"},{"alias_kind":"pith_short_8","alias_value":"4OYOAVSF","created_at":"2026-07-05T01:46:05.492405+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2303.08112","citing_title":"Eliciting Latent Predictions from Transformers with the Tuned Lens","ref_index":95,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3","json":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3.json","graph_json":"https://pith.science/api/pith-number/4OYOAVSFGG5O7C7LMVVHFNXRE3/graph.json","events_json":"https://pith.science/api/pith-number/4OYOAVSFGG5O7C7LMVVHFNXRE3/events.json","paper":"https://pith.science/paper/4OYOAVSF"},"agent_actions":{"view_html":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3","download_json":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3.json","view_paper":"https://pith.science/paper/4OYOAVSF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2010.13369&json=true","fetch_graph":"https://pith.science/api/pith-number/4OYOAVSFGG5O7C7LMVVHFNXRE3/graph.json","fetch_events":"https://pith.science/api/pith-number/4OYOAVSFGG5O7C7LMVVHFNXRE3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3/action/storage_attestation","attest_author":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3/action/author_attestation","sign_citation":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3/action/citation_signature","submit_replication":"https://pith.science/pith/4OYOAVSFGG5O7C7LMVVHFNXRE3/action/replication_record"}},"created_at":"2026-07-05T01:46:05.492405+00:00","updated_at":"2026-07-05T01:46:05.492405+00:00"}