{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:TJINR2UM473GKOOCFN7BNUB36Y","short_pith_number":"pith:TJINR2UM","schema_version":"1.0","canonical_sha256":"9a50d8ea8ce7f66539c22b7e16d03bf63d04bacd1a55a99da09d156ff5097ba3","source":{"kind":"arxiv","id":"2401.15024","version":2},"attestation_state":"computed","paper":{"title":"SliceGPT: Compress Large Language Models by Deleting Rows and Columns","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"James Hensman, Marcelo Gennari do Nascimento, Maximilian L. Croci, Saleh Ashkboos, Torsten Hoefler","submitted_at":"2024-01-26T17:35:45Z","abstract_excerpt":"Large language models have become the cornerstone of natural language processing, but their use comes with substantial costs in terms of compute and memory resources. Sparsification provides a solution to alleviate these resource constraints, and recent works have shown that trained models can be sparsified post-hoc. Existing sparsification techniques face challenges as they need additional data structures and offer constrained speedup with current hardware. In this paper we present SliceGPT, a new post-training sparsification scheme which replaces each weight matrix with a smaller (dense) mat"},"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":"2401.15024","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-01-26T17:35:45Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"414b3e177875e87a5ed4d21a357b64007bed53112557fa513f543a1b9ce15027","abstract_canon_sha256":"d07799cd327600733494c1fe4c77bd4bb055e422c9d736eaa64cf64d0bf87f24"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:43:14.464649Z","signature_b64":"10zlO1ggjaAlTVuibEcR8b7B/xNXaHM/VC2ngbqEf+YSH2ENZlAyhW9DoLa0dpGPpRF+N0r/IEFQWKSlcDGnAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a50d8ea8ce7f66539c22b7e16d03bf63d04bacd1a55a99da09d156ff5097ba3","last_reissued_at":"2026-07-05T07:43:14.464110Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:43:14.464110Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SliceGPT: Compress Large Language Models by Deleting Rows and Columns","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"James Hensman, Marcelo Gennari do Nascimento, Maximilian L. Croci, Saleh Ashkboos, Torsten Hoefler","submitted_at":"2024-01-26T17:35:45Z","abstract_excerpt":"Large language models have become the cornerstone of natural language processing, but their use comes with substantial costs in terms of compute and memory resources. Sparsification provides a solution to alleviate these resource constraints, and recent works have shown that trained models can be sparsified post-hoc. Existing sparsification techniques face challenges as they need additional data structures and offer constrained speedup with current hardware. In this paper we present SliceGPT, a new post-training sparsification scheme which replaces each weight matrix with a smaller (dense) mat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.15024","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/2401.15024/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":"2401.15024","created_at":"2026-07-05T07:43:14.464179+00:00"},{"alias_kind":"arxiv_version","alias_value":"2401.15024v2","created_at":"2026-07-05T07:43:14.464179+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.15024","created_at":"2026-07-05T07:43:14.464179+00:00"},{"alias_kind":"pith_short_12","alias_value":"TJINR2UM473G","created_at":"2026-07-05T07:43:14.464179+00:00"},{"alias_kind":"pith_short_16","alias_value":"TJINR2UM473GKOOC","created_at":"2026-07-05T07:43:14.464179+00:00"},{"alias_kind":"pith_short_8","alias_value":"TJINR2UM","created_at":"2026-07-05T07:43:14.464179+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":31,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07557","citing_title":"PALS: Percentile-Aware Layerwise Sparsity for LLM Pruning","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23607","citing_title":"Scaling Linear Mode Connectivity and Merging to Billion Parameter Pretrained Transformers","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23406","citing_title":"HyperQuant: A Rate-Distortion-Optimal Quantization Pipeline for Large Language and Diffusion Models","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.23670","citing_title":"Tapered Language Models","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19993","citing_title":"Activation- and Influence-Aware Ranks (AIR): Function-Preserving SVD Compression for LLMs","ref_index":166,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03465","citing_title":"Rethinking the Role of Tensor Decompositions in Post-Training LLM Compression","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01850","citing_title":"Does Compression Preserve Uncertainty? A Unified Benchmark for Quantized and Sparse LLMs via Conformal Prediction","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01666","citing_title":"DOT-MoE: Differentiable Optimal Transport for MoEfication","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01544","citing_title":"CRePE: Convolution-aware Relative Importance in Post-training Pruning with Efficient Search","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27743","citing_title":"End-to-End Dynamic Sparsity for Resource-Adaptive LLM Inference","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25344","citing_title":"A Hamiltonian-Inspired Local-Operator Ansatz for Slimming Large Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00523","citing_title":"ProactiveLLM: Learning Active Interaction for Streaming Large Language Models","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26861","citing_title":"Cascaded Multi-Granularity Pruning for On-Device LLM Inference in Industrial IoT","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2502.04416","citing_title":"Analytical FFN-to-MoE Restructuring via Activation Pattern Analysis","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2505.17138","citing_title":"RAP: Runtime Adaptive Pruning for LLM Inference","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14738","citing_title":"TAPIOCA: Why Task- Aware Pruning Improves OOD model Capability","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08738","citing_title":"SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17985","citing_title":"SAFE-SVD: Sensitivity-Aware Fidelity-Enforcing SVD for Physics Foundation Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18331","citing_title":"Prune, Update and Trim: Robust Structured Pruning for Large Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2506.12876","citing_title":"MaskPro: Linear-Space Probabilistic Learning for Strict (N:M)-Sparsity on LLMs","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2512.22671","citing_title":"Fragile Knowledge, Robust Instruction-Following: The Width Pruning Dichotomy in Llama-3.2","ref_index":2,"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":1,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":179,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03298","citing_title":"ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08568","citing_title":"Different Prompts, Different Ranks: Prompt-aware Dynamic Rank Selection for SVD-based LLM Compression","ref_index":57,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y","json":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y.json","graph_json":"https://pith.science/api/pith-number/TJINR2UM473GKOOCFN7BNUB36Y/graph.json","events_json":"https://pith.science/api/pith-number/TJINR2UM473GKOOCFN7BNUB36Y/events.json","paper":"https://pith.science/paper/TJINR2UM"},"agent_actions":{"view_html":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y","download_json":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y.json","view_paper":"https://pith.science/paper/TJINR2UM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2401.15024&json=true","fetch_graph":"https://pith.science/api/pith-number/TJINR2UM473GKOOCFN7BNUB36Y/graph.json","fetch_events":"https://pith.science/api/pith-number/TJINR2UM473GKOOCFN7BNUB36Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y/action/storage_attestation","attest_author":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y/action/author_attestation","sign_citation":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y/action/citation_signature","submit_replication":"https://pith.science/pith/TJINR2UM473GKOOCFN7BNUB36Y/action/replication_record"}},"created_at":"2026-07-05T07:43:14.464179+00:00","updated_at":"2026-07-05T07:43:14.464179+00:00"}