{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:BHESSHYMS6IUARA4OLT6NMJAGU","short_pith_number":"pith:BHESSHYM","canonical_record":{"source":{"id":"2302.03773","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-07T22:05:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce2666fd913c10ecb53f5df302a1353b7b30036454365d6e9bba4e5138b013cd","abstract_canon_sha256":"d7748f5cd5b03bb2e8d008e6244074abf4c0219148d5860d6f2d9baa191caa7d"},"schema_version":"1.0"},"canonical_sha256":"09c9291f0c979140441c72e7e6b120353e3e496404f374a0e98048b14ead938f","source":{"kind":"arxiv","id":"2302.03773","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.03773","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"arxiv_version","alias_value":"2302.03773v1","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.03773","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_12","alias_value":"BHESSHYMS6IU","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_16","alias_value":"BHESSHYMS6IUARA4","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_8","alias_value":"BHESSHYM","created_at":"2026-07-05T05:39:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:BHESSHYMS6IUARA4OLT6NMJAGU","target":"record","payload":{"canonical_record":{"source":{"id":"2302.03773","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-07T22:05:55Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"ce2666fd913c10ecb53f5df302a1353b7b30036454365d6e9bba4e5138b013cd","abstract_canon_sha256":"d7748f5cd5b03bb2e8d008e6244074abf4c0219148d5860d6f2d9baa191caa7d"},"schema_version":"1.0"},"canonical_sha256":"09c9291f0c979140441c72e7e6b120353e3e496404f374a0e98048b14ead938f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:39:57.503476Z","signature_b64":"SRSVQ+AMNObMLH16uEmHPfA7MPhP2kzyUbLrgE+lPT/EIvV9fNO5BVL/JFj9HjRtAvriVAoovM2d5+bOjL4uAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"09c9291f0c979140441c72e7e6b120353e3e496404f374a0e98048b14ead938f","last_reissued_at":"2026-07-05T05:39:57.502983Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:39:57.502983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2302.03773","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:39:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tZegZrkSRcK0iRVwlEDjIdX2BZYMJ/MJIb7JKYnnNFoerCo9/qBvOwPLaMXjeaXifkkh0KX7dgHWOLh5i92QCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T05:12:05.924570Z"},"content_sha256":"1ab5314c0e73805c66d5581f2caaa6bd09dbf42ae23988bfc2a1238496124221","schema_version":"1.0","event_id":"sha256:1ab5314c0e73805c66d5581f2caaa6bd09dbf42ae23988bfc2a1238496124221"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:BHESSHYMS6IUARA4OLT6NMJAGU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"What Matters In The Structured Pruning of Generative Language Models?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Michael Santacroce, Yelong Shen, Yuanzhi Li, Zixin Wen","submitted_at":"2023-02-07T22:05:55Z","abstract_excerpt":"Auto-regressive large language models such as GPT-3 require enormous computational resources to use. Traditionally, structured pruning methods are employed to reduce resource usage. However, their application to and efficacy for generative language models is heavily under-explored. In this paper we conduct an comprehensive evaluation of common structured pruning methods, including magnitude, random, and movement pruning on the feed-forward layers in GPT-type models. Unexpectedly, random pruning results in performance that is comparable to the best established methods, across multiple natural l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.03773","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/2302.03773/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":"2026-07-05T05:39:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eMLeCAxhMfZzWyk3NsGmEjs0pb+J4N3Hj+XeQJ9WnY02alJNcvUMhkxQ8TNmBQ9+64P2qiIzcO4rX+bkIeWfAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T05:12:05.925734Z"},"content_sha256":"c046fd36ff317cc5cd790c2abf19de7ea8db08429ca7a11485c9aa574254ca83","schema_version":"1.0","event_id":"sha256:c046fd36ff317cc5cd790c2abf19de7ea8db08429ca7a11485c9aa574254ca83"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BHESSHYMS6IUARA4OLT6NMJAGU/bundle.json","state_url":"https://pith.science/pith/BHESSHYMS6IUARA4OLT6NMJAGU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BHESSHYMS6IUARA4OLT6NMJAGU/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":"2026-08-11T05:12:05Z","links":{"resolver":"https://pith.science/pith/BHESSHYMS6IUARA4OLT6NMJAGU","bundle":"https://pith.science/pith/BHESSHYMS6IUARA4OLT6NMJAGU/bundle.json","state":"https://pith.science/pith/BHESSHYMS6IUARA4OLT6NMJAGU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BHESSHYMS6IUARA4OLT6NMJAGU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:BHESSHYMS6IUARA4OLT6NMJAGU","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":"d7748f5cd5b03bb2e8d008e6244074abf4c0219148d5860d6f2d9baa191caa7d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-07T22:05:55Z","title_canon_sha256":"ce2666fd913c10ecb53f5df302a1353b7b30036454365d6e9bba4e5138b013cd"},"schema_version":"1.0","source":{"id":"2302.03773","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2302.03773","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"arxiv_version","alias_value":"2302.03773v1","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2302.03773","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_12","alias_value":"BHESSHYMS6IU","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_16","alias_value":"BHESSHYMS6IUARA4","created_at":"2026-07-05T05:39:57Z"},{"alias_kind":"pith_short_8","alias_value":"BHESSHYM","created_at":"2026-07-05T05:39:57Z"}],"graph_snapshots":[{"event_id":"sha256:c046fd36ff317cc5cd790c2abf19de7ea8db08429ca7a11485c9aa574254ca83","target":"graph","created_at":"2026-07-05T05:39:57Z","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/2302.03773/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Auto-regressive large language models such as GPT-3 require enormous computational resources to use. Traditionally, structured pruning methods are employed to reduce resource usage. However, their application to and efficacy for generative language models is heavily under-explored. In this paper we conduct an comprehensive evaluation of common structured pruning methods, including magnitude, random, and movement pruning on the feed-forward layers in GPT-type models. Unexpectedly, random pruning results in performance that is comparable to the best established methods, across multiple natural l","authors_text":"Michael Santacroce, Yelong Shen, Yuanzhi Li, Zixin Wen","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-07T22:05:55Z","title":"What Matters In The Structured Pruning of Generative Language Models?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2302.03773","kind":"arxiv","version":1},"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":"sha256:1ab5314c0e73805c66d5581f2caaa6bd09dbf42ae23988bfc2a1238496124221","target":"record","created_at":"2026-07-05T05:39:57Z","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":"d7748f5cd5b03bb2e8d008e6244074abf4c0219148d5860d6f2d9baa191caa7d","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-02-07T22:05:55Z","title_canon_sha256":"ce2666fd913c10ecb53f5df302a1353b7b30036454365d6e9bba4e5138b013cd"},"schema_version":"1.0","source":{"id":"2302.03773","kind":"arxiv","version":1}},"canonical_sha256":"09c9291f0c979140441c72e7e6b120353e3e496404f374a0e98048b14ead938f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"09c9291f0c979140441c72e7e6b120353e3e496404f374a0e98048b14ead938f","first_computed_at":"2026-07-05T05:39:57.502983Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:39:57.502983Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SRSVQ+AMNObMLH16uEmHPfA7MPhP2kzyUbLrgE+lPT/EIvV9fNO5BVL/JFj9HjRtAvriVAoovM2d5+bOjL4uAg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:39:57.503476Z","signed_message":"canonical_sha256_bytes"},"source_id":"2302.03773","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1ab5314c0e73805c66d5581f2caaa6bd09dbf42ae23988bfc2a1238496124221","sha256:c046fd36ff317cc5cd790c2abf19de7ea8db08429ca7a11485c9aa574254ca83"],"state_sha256":"f292d1c7b08f3ddd7bb54e64dc688b3923081ab03b3045e09b66dbf643c93398"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dJJe/MJ3cCwGHiHd/RXmw3PKM+ogS7E7ORCMA4X1GaBun6SIkXaAKUUeRf75gtXUVPXXv7oGAeHBJPgkwkU8Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T05:12:05.933193Z","bundle_sha256":"defe951272782c0e65dafa98a96162445c58580452ae1897aaf98d4d350d4bf3"}}