{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OMHV5D2BTCVJPNDEZXDDKAH3OZ","short_pith_number":"pith:OMHV5D2B","schema_version":"1.0","canonical_sha256":"730f5e8f4198aa97b464cdc63500fb767b547da2b4d428a4fa19f1e523cb95cf","source":{"kind":"arxiv","id":"2502.07780","version":3},"attestation_state":"computed","paper":{"title":"DarwinLM: Evolutionary Structured Pruning of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Dan Alistarh, Eldar Kurtic, Oliver Sieberling, Shengkun Tang, Zhiqiang Shen","submitted_at":"2025-02-11T18:59:35Z","abstract_excerpt":"Large Language Models (LLMs) have achieved significant success across various NLP tasks. However, their massive computational costs limit their widespread use, particularly in real-time applications. Structured pruning offers an effective solution by compressing models and directly providing end-to-end speed improvements, regardless of the hardware environment. Meanwhile, different components of the model exhibit varying sensitivities towards pruning, calling for non-uniform model compression. However, a pruning method should not only identify a capable substructure, but also account for post-"},"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":"2502.07780","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-11T18:59:35Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"85f61680d70468308abce1ae0ac8bca1cf743b151e70657d53aff5af28fc89a2","abstract_canon_sha256":"75a7c652faa677cc79f30fe0f8aab8916eaf44a4ae394987039bc2fd9cc6e869"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:30.863887Z","signature_b64":"ONgkCQWGoGBDu7xspNV0cMVybV/1gE/cCe/JByvZs3VSiTIsJeEHc3BZDIm/3PwccazwL+NETjJyEnAoFOyNDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"730f5e8f4198aa97b464cdc63500fb767b547da2b4d428a4fa19f1e523cb95cf","last_reissued_at":"2026-07-05T10:24:30.863145Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:30.863145Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DarwinLM: Evolutionary Structured Pruning of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Dan Alistarh, Eldar Kurtic, Oliver Sieberling, Shengkun Tang, Zhiqiang Shen","submitted_at":"2025-02-11T18:59:35Z","abstract_excerpt":"Large Language Models (LLMs) have achieved significant success across various NLP tasks. However, their massive computational costs limit their widespread use, particularly in real-time applications. Structured pruning offers an effective solution by compressing models and directly providing end-to-end speed improvements, regardless of the hardware environment. Meanwhile, different components of the model exhibit varying sensitivities towards pruning, calling for non-uniform model compression. However, a pruning method should not only identify a capable substructure, but also account for post-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07780","kind":"arxiv","version":3},"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/2502.07780/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":"2502.07780","created_at":"2026-07-05T10:24:30.863240+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07780v3","created_at":"2026-07-05T10:24:30.863240+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07780","created_at":"2026-07-05T10:24:30.863240+00:00"},{"alias_kind":"pith_short_12","alias_value":"OMHV5D2BTCVJ","created_at":"2026-07-05T10:24:30.863240+00:00"},{"alias_kind":"pith_short_16","alias_value":"OMHV5D2BTCVJPNDE","created_at":"2026-07-05T10:24:30.863240+00:00"},{"alias_kind":"pith_short_8","alias_value":"OMHV5D2B","created_at":"2026-07-05T10:24:30.863240+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":7,"sample":[{"citing_arxiv_id":"2606.02559","citing_title":"From Layers to Submodules: Rethinking Granularity in Replacement-Based LLM Compression","ref_index":118,"is_internal_anchor":true},{"citing_arxiv_id":"2605.14738","citing_title":"TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08738","citing_title":"SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2603.06003","citing_title":"EvoESAP: Non-Uniform Expert Pruning for Sparse MoE","ref_index":52,"is_internal_anchor":true},{"citing_arxiv_id":"2604.24938","citing_title":"Rethinking Layer Redundancy in Large Language Models: Calibration Objectives and Search for Depth Pruning","ref_index":19,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08738","citing_title":"SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training","ref_index":59,"is_internal_anchor":true},{"citing_arxiv_id":"2604.24938","citing_title":"Rethinking Layer Redundancy in Large Language Models: Calibration Objectives and Search for Depth Pruning","ref_index":19,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ","json":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ.json","graph_json":"https://pith.science/api/pith-number/OMHV5D2BTCVJPNDEZXDDKAH3OZ/graph.json","events_json":"https://pith.science/api/pith-number/OMHV5D2BTCVJPNDEZXDDKAH3OZ/events.json","paper":"https://pith.science/paper/OMHV5D2B"},"agent_actions":{"view_html":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ","download_json":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ.json","view_paper":"https://pith.science/paper/OMHV5D2B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07780&json=true","fetch_graph":"https://pith.science/api/pith-number/OMHV5D2BTCVJPNDEZXDDKAH3OZ/graph.json","fetch_events":"https://pith.science/api/pith-number/OMHV5D2BTCVJPNDEZXDDKAH3OZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ/action/storage_attestation","attest_author":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ/action/author_attestation","sign_citation":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ/action/citation_signature","submit_replication":"https://pith.science/pith/OMHV5D2BTCVJPNDEZXDDKAH3OZ/action/replication_record"}},"created_at":"2026-07-05T10:24:30.863240+00:00","updated_at":"2026-07-05T10:24:30.863240+00:00"}