{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:TS3HO6W7QLDHEO7UE325BIQNLE","short_pith_number":"pith:TS3HO6W7","schema_version":"1.0","canonical_sha256":"9cb6777adf82c6723bf426f5d0a20d5914f88badc73231742eceadc273120f8d","source":{"kind":"arxiv","id":"2208.11580","version":2},"attestation_state":"computed","paper":{"title":"Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dan Alistarh, Elias Frantar, Sidak Pal Singh","submitted_at":"2022-08-24T14:33:35Z","abstract_excerpt":"We consider the problem of model compression for deep neural networks (DNNs) in the challenging one-shot/post-training setting, in which we are given an accurate trained model, and must compress it without any retraining, based only on a small amount of calibration input data. This problem has become popular in view of the emerging software and hardware support for executing models compressed via pruning and/or quantization with speedup, and well-performing solutions have been proposed independently for both compression approaches. In this paper, we introduce a new compression framework which "},"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":"2208.11580","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-08-24T14:33:35Z","cross_cats_sorted":[],"title_canon_sha256":"17bfc410a4ed8df0b0188ab06e5724d0b968c747440e226879e8ef8bbd0f53ad","abstract_canon_sha256":"b47d03ae9b59db324d11f0d2d10f25efb20410d7d42e4cd1cf65fe2dab9a88bf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:31:16.519243Z","signature_b64":"KAdrrSIc3fCOYxb4LFx2p4HbIR11bOIlsgeZhYOfCTGOyMuDw0iCMmjeMqNNd3oaO1pguUz0baO5HBoQ/GqRBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9cb6777adf82c6723bf426f5d0a20d5914f88badc73231742eceadc273120f8d","last_reissued_at":"2026-07-05T05:31:16.518772Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:31:16.518772Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Optimal Brain Compression: A Framework for Accurate Post-Training Quantization and Pruning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dan Alistarh, Elias Frantar, Sidak Pal Singh","submitted_at":"2022-08-24T14:33:35Z","abstract_excerpt":"We consider the problem of model compression for deep neural networks (DNNs) in the challenging one-shot/post-training setting, in which we are given an accurate trained model, and must compress it without any retraining, based only on a small amount of calibration input data. This problem has become popular in view of the emerging software and hardware support for executing models compressed via pruning and/or quantization with speedup, and well-performing solutions have been proposed independently for both compression approaches. In this paper, we introduce a new compression framework which "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.11580","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/2208.11580/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":"2208.11580","created_at":"2026-07-05T05:31:16.518829+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.11580v2","created_at":"2026-07-05T05:31:16.518829+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.11580","created_at":"2026-07-05T05:31:16.518829+00:00"},{"alias_kind":"pith_short_12","alias_value":"TS3HO6W7QLDH","created_at":"2026-07-05T05:31:16.518829+00:00"},{"alias_kind":"pith_short_16","alias_value":"TS3HO6W7QLDHEO7U","created_at":"2026-07-05T05:31:16.518829+00:00"},{"alias_kind":"pith_short_8","alias_value":"TS3HO6W7","created_at":"2026-07-05T05:31:16.518829+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.24754","citing_title":"Motion-Compensated Weight Compression","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2601.02455","citing_title":"Diagnostic-Driven Layer-Wise Compensation for Post-Training Quantization of Encoder-Decoder ASR Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05243","citing_title":"CORP: Closed-Form One-shot Representation-Preserving Structured Pruning for Transformers","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2210.17323","citing_title":"GPTQ: Accurate Post-Training Quantization for Generative Pre-trained Transformers","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE","json":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE.json","graph_json":"https://pith.science/api/pith-number/TS3HO6W7QLDHEO7UE325BIQNLE/graph.json","events_json":"https://pith.science/api/pith-number/TS3HO6W7QLDHEO7UE325BIQNLE/events.json","paper":"https://pith.science/paper/TS3HO6W7"},"agent_actions":{"view_html":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE","download_json":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE.json","view_paper":"https://pith.science/paper/TS3HO6W7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.11580&json=true","fetch_graph":"https://pith.science/api/pith-number/TS3HO6W7QLDHEO7UE325BIQNLE/graph.json","fetch_events":"https://pith.science/api/pith-number/TS3HO6W7QLDHEO7UE325BIQNLE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE/action/storage_attestation","attest_author":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE/action/author_attestation","sign_citation":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE/action/citation_signature","submit_replication":"https://pith.science/pith/TS3HO6W7QLDHEO7UE325BIQNLE/action/replication_record"}},"created_at":"2026-07-05T05:31:16.518829+00:00","updated_at":"2026-07-05T05:31:16.518829+00:00"}