{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FQLRKD3LEAECLYPMIRUQ7AC74D","short_pith_number":"pith:FQLRKD3L","schema_version":"1.0","canonical_sha256":"2c17150f6b200825e1ec44690f805fe0fbf014bbcde604728d11a0e31ebfee53","source":{"kind":"arxiv","id":"2402.09025","version":6},"attestation_state":"computed","paper":{"title":"SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hyungjun Kim, Jae-Joon Kim, Jiwon Song, Kyungseok Oh, Taesu Kim, Yulhwa Kim","submitted_at":"2024-02-14T09:01:13Z","abstract_excerpt":"Large language models (LLMs) have proven to be highly effective across various natural language processing tasks. However, their large number of parameters poses significant challenges for practical deployment. Pruning, a technique aimed at reducing the size and complexity of LLMs, offers a potential solution by removing redundant components from the network. Despite the promise of pruning, existing methods often struggle to achieve substantial end-to-end LLM inference speedup. In this paper, we introduce SLEB, a novel approach designed to streamline LLMs by eliminating redundant transformer b"},"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":"2402.09025","kind":"arxiv","version":6},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-14T09:01:13Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"b88fdcb82721b0a8988eedcb066339ca0102663ccc76a28f79f61ee0e5ceaed4","abstract_canon_sha256":"533c6a166a483b8ee634e345b56b1c4ec8375d436c0a708fc055c47636699b49"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:31.230772Z","signature_b64":"JD7DgvDGNaVChz6Ylrz3MhyBOslZVI1RMhmM2O9/YJ1WE8EyJ8tWxPkH1ZNQIgaxiEPQdmez3G1gwnstqrEmCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c17150f6b200825e1ec44690f805fe0fbf014bbcde604728d11a0e31ebfee53","last_reissued_at":"2026-07-05T09:48:31.230289Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:31.230289Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SLEB: Streamlining LLMs through Redundancy Verification and Elimination of Transformer Blocks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Hyungjun Kim, Jae-Joon Kim, Jiwon Song, Kyungseok Oh, Taesu Kim, Yulhwa Kim","submitted_at":"2024-02-14T09:01:13Z","abstract_excerpt":"Large language models (LLMs) have proven to be highly effective across various natural language processing tasks. However, their large number of parameters poses significant challenges for practical deployment. Pruning, a technique aimed at reducing the size and complexity of LLMs, offers a potential solution by removing redundant components from the network. Despite the promise of pruning, existing methods often struggle to achieve substantial end-to-end LLM inference speedup. In this paper, we introduce SLEB, a novel approach designed to streamline LLMs by eliminating redundant transformer b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.09025","kind":"arxiv","version":6},"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/2402.09025/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":"2402.09025","created_at":"2026-07-05T09:48:31.230355+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.09025v6","created_at":"2026-07-05T09:48:31.230355+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.09025","created_at":"2026-07-05T09:48:31.230355+00:00"},{"alias_kind":"pith_short_12","alias_value":"FQLRKD3LEAEC","created_at":"2026-07-05T09:48:31.230355+00:00"},{"alias_kind":"pith_short_16","alias_value":"FQLRKD3LEAECLYPM","created_at":"2026-07-05T09:48:31.230355+00:00"},{"alias_kind":"pith_short_8","alias_value":"FQLRKD3L","created_at":"2026-07-05T09:48:31.230355+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01666","citing_title":"DOT-MoE: Differentiable Optimal Transport for MoEfication","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14738","citing_title":"TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16234","citing_title":"No Free Swap: Protocol-Dependent Layer Redundancy in Transformers","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2602.01997","citing_title":"On the Limits of Layer Pruning for Generative Reasoning in Large Language Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07271","citing_title":"Understanding Performance Collapse in Layer-Pruned Large Language Models via Decision Representation Transitions","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D","json":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D.json","graph_json":"https://pith.science/api/pith-number/FQLRKD3LEAECLYPMIRUQ7AC74D/graph.json","events_json":"https://pith.science/api/pith-number/FQLRKD3LEAECLYPMIRUQ7AC74D/events.json","paper":"https://pith.science/paper/FQLRKD3L"},"agent_actions":{"view_html":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D","download_json":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D.json","view_paper":"https://pith.science/paper/FQLRKD3L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.09025&json=true","fetch_graph":"https://pith.science/api/pith-number/FQLRKD3LEAECLYPMIRUQ7AC74D/graph.json","fetch_events":"https://pith.science/api/pith-number/FQLRKD3LEAECLYPMIRUQ7AC74D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D/action/storage_attestation","attest_author":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D/action/author_attestation","sign_citation":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D/action/citation_signature","submit_replication":"https://pith.science/pith/FQLRKD3LEAECLYPMIRUQ7AC74D/action/replication_record"}},"created_at":"2026-07-05T09:48:31.230355+00:00","updated_at":"2026-07-05T09:48:31.230355+00:00"}