{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:6WRAP3APC2ZAZZJZWXHJ6FJGOR","short_pith_number":"pith:6WRAP3AP","schema_version":"1.0","canonical_sha256":"f5a207ec0f16b20ce539b5ce9f152674736cfeac84592725edecc863b5af0f40","source":{"kind":"arxiv","id":"2404.09497","version":1},"attestation_state":"computed","paper":{"title":"Towards Efficient SRAM-PIM Architecture Design by Exploiting Unstructured Bit-Level Sparsity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Bonan Yan, Cenlin Duan, Jianlei Yang, Weisheng Zhao, Xiaolin He, Xiaotao Jia, Xueyan Wang, Yikun Wang, Yingjie Qi, Yiou Wang","submitted_at":"2024-04-15T06:34:17Z","abstract_excerpt":"Bit-level sparsity in neural network models harbors immense untapped potential. Eliminating redundant calculations of randomly distributed zero-bits significantly boosts computational efficiency. Yet, traditional digital SRAM-PIM architecture, limited by rigid crossbar architecture, struggles to effectively exploit this unstructured sparsity. To address this challenge, we propose Dyadic Block PIM (DB-PIM), a groundbreaking algorithm-architecture co-design framework. First, we propose an algorithm coupled with a distinctive sparsity pattern, termed a dyadic block (DB), that preserves the random"},"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":"2404.09497","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-04-15T06:34:17Z","cross_cats_sorted":[],"title_canon_sha256":"c4c2d577846c63e272278fde78ee9edf329eb1fb8ccbef55cdbd48cc0255e3bb","abstract_canon_sha256":"174d870ca0a810578b6db7ef71e16c348fdf33387019e91a12a9163698fce8ea"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:08:07.302803Z","signature_b64":"ZNsHbf9cAwcORKdLI2Vuq+yrPmy/BEdUYGnBpV8R5CG4+MYthWr5fJdD/AhIpz5w4ey3kJt6bPGvTwVrs+mrCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5a207ec0f16b20ce539b5ce9f152674736cfeac84592725edecc863b5af0f40","last_reissued_at":"2026-07-05T08:08:07.302271Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:08:07.302271Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Towards Efficient SRAM-PIM Architecture Design by Exploiting Unstructured Bit-Level Sparsity","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Bonan Yan, Cenlin Duan, Jianlei Yang, Weisheng Zhao, Xiaolin He, Xiaotao Jia, Xueyan Wang, Yikun Wang, Yingjie Qi, Yiou Wang","submitted_at":"2024-04-15T06:34:17Z","abstract_excerpt":"Bit-level sparsity in neural network models harbors immense untapped potential. Eliminating redundant calculations of randomly distributed zero-bits significantly boosts computational efficiency. Yet, traditional digital SRAM-PIM architecture, limited by rigid crossbar architecture, struggles to effectively exploit this unstructured sparsity. To address this challenge, we propose Dyadic Block PIM (DB-PIM), a groundbreaking algorithm-architecture co-design framework. First, we propose an algorithm coupled with a distinctive sparsity pattern, termed a dyadic block (DB), that preserves the random"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.09497","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/2404.09497/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":"2404.09497","created_at":"2026-07-05T08:08:07.302376+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.09497v1","created_at":"2026-07-05T08:08:07.302376+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.09497","created_at":"2026-07-05T08:08:07.302376+00:00"},{"alias_kind":"pith_short_12","alias_value":"6WRAP3APC2ZA","created_at":"2026-07-05T08:08:07.302376+00:00"},{"alias_kind":"pith_short_16","alias_value":"6WRAP3APC2ZAZZJZ","created_at":"2026-07-05T08:08:07.302376+00:00"},{"alias_kind":"pith_short_8","alias_value":"6WRAP3AP","created_at":"2026-07-05T08:08:07.302376+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2504.14365","citing_title":"Accelerating LLM Inference with Flexible N:M Sparsity via A Fully Digital Compute-in-Memory Accelerator","ref_index":7,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR","json":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR.json","graph_json":"https://pith.science/api/pith-number/6WRAP3APC2ZAZZJZWXHJ6FJGOR/graph.json","events_json":"https://pith.science/api/pith-number/6WRAP3APC2ZAZZJZWXHJ6FJGOR/events.json","paper":"https://pith.science/paper/6WRAP3AP"},"agent_actions":{"view_html":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR","download_json":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR.json","view_paper":"https://pith.science/paper/6WRAP3AP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.09497&json=true","fetch_graph":"https://pith.science/api/pith-number/6WRAP3APC2ZAZZJZWXHJ6FJGOR/graph.json","fetch_events":"https://pith.science/api/pith-number/6WRAP3APC2ZAZZJZWXHJ6FJGOR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR/action/storage_attestation","attest_author":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR/action/author_attestation","sign_citation":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR/action/citation_signature","submit_replication":"https://pith.science/pith/6WRAP3APC2ZAZZJZWXHJ6FJGOR/action/replication_record"}},"created_at":"2026-07-05T08:08:07.302376+00:00","updated_at":"2026-07-05T08:08:07.302376+00:00"}