{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:TR33AAXEHRVXKAIJOYJCYGHOP5","short_pith_number":"pith:TR33AAXE","schema_version":"1.0","canonical_sha256":"9c77b002e43c6b75010976122c18ee7f6ed254e614798d7ffd18745b227cf974","source":{"kind":"arxiv","id":"2104.08378","version":1},"attestation_state":"computed","paper":{"title":"Accelerating Sparse Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.LG","authors_text":"Asit Mishra, Chong Yu, Darko Stosic, Dusan Stosic, Ganesh Venkatesh, Jeff Pool, Jorge Albericio Latorre, Paulius Micikevicius","submitted_at":"2021-04-16T21:27:32Z","abstract_excerpt":"As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in parameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsi"},"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":"2104.08378","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-04-16T21:27:32Z","cross_cats_sorted":["cs.AI","cs.AR"],"title_canon_sha256":"88a2137386afb6c3512810c3b60604febbe56fc7fff0e44927921ba1a2b5494f","abstract_canon_sha256":"85b65facc71092ce4c2fc63b28ada789154aeb25a036126599ac0514669b238f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:32:51.308653Z","signature_b64":"eE0bBK0tKhr/V53/915BN8lNUVxP3CZ8IwyS8EnUugFlbFYtepRqM3XvPZM0BJwOdeSr+dHy4p3SapoDh7fzCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c77b002e43c6b75010976122c18ee7f6ed254e614798d7ffd18745b227cf974","last_reissued_at":"2026-07-05T02:32:51.308156Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:32:51.308156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Sparse Deep Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.AR"],"primary_cat":"cs.LG","authors_text":"Asit Mishra, Chong Yu, Darko Stosic, Dusan Stosic, Ganesh Venkatesh, Jeff Pool, Jorge Albericio Latorre, Paulius Micikevicius","submitted_at":"2021-04-16T21:27:32Z","abstract_excerpt":"As neural network model sizes have dramatically increased, so has the interest in various techniques to reduce their parameter counts and accelerate their execution. An active area of research in this field is sparsity - encouraging zero values in parameters that can then be discarded from storage or computations. While most research focuses on high levels of sparsity, there are challenges in universally maintaining model accuracy as well as achieving significant speedups over modern matrix-math hardware. To make sparsity adoption practical, the NVIDIA Ampere GPU architecture introduces sparsi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.08378","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/2104.08378/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":"2104.08378","created_at":"2026-07-05T02:32:51.308214+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.08378v1","created_at":"2026-07-05T02:32:51.308214+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.08378","created_at":"2026-07-05T02:32:51.308214+00:00"},{"alias_kind":"pith_short_12","alias_value":"TR33AAXEHRVX","created_at":"2026-07-05T02:32:51.308214+00:00"},{"alias_kind":"pith_short_16","alias_value":"TR33AAXEHRVXKAIJ","created_at":"2026-07-05T02:32:51.308214+00:00"},{"alias_kind":"pith_short_8","alias_value":"TR33AAXE","created_at":"2026-07-05T02:32:51.308214+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":15,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22296","citing_title":"SCENIC: Semantic-Conditioned Edge-Aware Neural Framework for Structured IoT Command Generation","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.14150","citing_title":"Small LLMs: Pruning vs. Training from Scratch","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00760","citing_title":"MosaicKV: Serving Long-Context LLM with Dynamic Two-D KV Cache Compression","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17289","citing_title":"LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.02608","citing_title":"Pruning Deep Neural Networks via the Marchenko--Pastur Distribution","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2606.14150","citing_title":"Small LLMs: Pruning vs. Training from Scratch","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10445","citing_title":"SpenseGPT: Practical One-shot Pruning Enabling Sparse and Dense GEMMs for LLM Inference","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21104","citing_title":"HORST: Composing Optimizer Geometries for Sparse Transformer Training","ref_index":249,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17289","citing_title":"LEAP: Learnable End-to-End Adaptive Pruning of Large Language Models","ref_index":8,"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":19,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03667","citing_title":"ELAS: Efficient Pre-Training of Low-Rank Large Language Models via 2:4 Activation Sparsity","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.06402","citing_title":"SparseForge: Efficient Semi-Structured LLM Sparsification via Annealing of Hessian-Guided Soft-Mask","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00171","citing_title":"Adaptive Norm-Based Regularization for Neural Networks","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04493","citing_title":"SLaB: Sparse-Lowrank-Binary Decomposition for Efficient Large Language Models","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16864","citing_title":"HieraSparse: Hierarchical Semi-Structured Sparse KV Attention","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5","json":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5.json","graph_json":"https://pith.science/api/pith-number/TR33AAXEHRVXKAIJOYJCYGHOP5/graph.json","events_json":"https://pith.science/api/pith-number/TR33AAXEHRVXKAIJOYJCYGHOP5/events.json","paper":"https://pith.science/paper/TR33AAXE"},"agent_actions":{"view_html":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5","download_json":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5.json","view_paper":"https://pith.science/paper/TR33AAXE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.08378&json=true","fetch_graph":"https://pith.science/api/pith-number/TR33AAXEHRVXKAIJOYJCYGHOP5/graph.json","fetch_events":"https://pith.science/api/pith-number/TR33AAXEHRVXKAIJOYJCYGHOP5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5/action/storage_attestation","attest_author":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5/action/author_attestation","sign_citation":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5/action/citation_signature","submit_replication":"https://pith.science/pith/TR33AAXEHRVXKAIJOYJCYGHOP5/action/replication_record"}},"created_at":"2026-07-05T02:32:51.308214+00:00","updated_at":"2026-07-05T02:32:51.308214+00:00"}