{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:TWMJCLWO6SBRQFP6KVSN3Z6TID","short_pith_number":"pith:TWMJCLWO","schema_version":"1.0","canonical_sha256":"9d98912ecef4831815fe5564dde7d340ed0190f07cb6b03a9fa7ec9eceb50b30","source":{"kind":"arxiv","id":"2509.10372","version":1},"attestation_state":"computed","paper":{"title":"MCBP: A Memory-Compute Efficient LLM Inference Accelerator Leveraging Bit-Slice-enabled Sparsity and Repetitiveness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Chao Li, Huizheng Wang, Jianxun Yang, Shaojun Wei, Shouyi Yin, Taiquan Wei, Yang Hu, Yang Wang, Yousheng Long, Zhiheng Yue, Zichuan Wang","submitted_at":"2025-09-12T16:05:27Z","abstract_excerpt":"Large language models (LLMs) face significant inference latency due to inefficiencies in GEMM operations, weight access, and KV cache access, especially in real-time scenarios. This highlights the need for a versatile compute-memory efficient accelerator. Unfortunately, existing Transformer accelerators struggle to address both aspects simultaneously, as they focus on value-level processing, missing fine-grained opportunities to optimize computation and memory collaboratively. This paper introduces MCBP, a bit-grained compute-memory efficient algorithm-hardware co-design that leverages bit-sli"},"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":"2509.10372","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-09-12T16:05:27Z","cross_cats_sorted":[],"title_canon_sha256":"2b335ce99102aacd925a06deeacb8dcdf3a9069708de61cc8b1c0d691cc89843","abstract_canon_sha256":"3bedb00ecfe6c311a6e6102f5e4966d4f594ae16bf2e955cf7a3b8aa428570bb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:08.682300Z","signature_b64":"r2Ic9FefczjoVrBxW7QZmoKVOHMpSBKm5iHK0F+jzLizrKEN3ef+mWua6Gemx1EtVIFrWoMvmxItJ+bjEKnhDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d98912ecef4831815fe5564dde7d340ed0190f07cb6b03a9fa7ec9eceb50b30","last_reissued_at":"2026-07-05T12:11:08.681744Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:08.681744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MCBP: A Memory-Compute Efficient LLM Inference Accelerator Leveraging Bit-Slice-enabled Sparsity and Repetitiveness","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AR","authors_text":"Chao Li, Huizheng Wang, Jianxun Yang, Shaojun Wei, Shouyi Yin, Taiquan Wei, Yang Hu, Yang Wang, Yousheng Long, Zhiheng Yue, Zichuan Wang","submitted_at":"2025-09-12T16:05:27Z","abstract_excerpt":"Large language models (LLMs) face significant inference latency due to inefficiencies in GEMM operations, weight access, and KV cache access, especially in real-time scenarios. This highlights the need for a versatile compute-memory efficient accelerator. Unfortunately, existing Transformer accelerators struggle to address both aspects simultaneously, as they focus on value-level processing, missing fine-grained opportunities to optimize computation and memory collaboratively. This paper introduces MCBP, a bit-grained compute-memory efficient algorithm-hardware co-design that leverages bit-sli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.10372","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/2509.10372/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":"2509.10372","created_at":"2026-07-05T12:11:08.681806+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.10372v1","created_at":"2026-07-05T12:11:08.681806+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.10372","created_at":"2026-07-05T12:11:08.681806+00:00"},{"alias_kind":"pith_short_12","alias_value":"TWMJCLWO6SBR","created_at":"2026-07-05T12:11:08.681806+00:00"},{"alias_kind":"pith_short_16","alias_value":"TWMJCLWO6SBRQFP6","created_at":"2026-07-05T12:11:08.681806+00:00"},{"alias_kind":"pith_short_8","alias_value":"TWMJCLWO","created_at":"2026-07-05T12:11:08.681806+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID","json":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID.json","graph_json":"https://pith.science/api/pith-number/TWMJCLWO6SBRQFP6KVSN3Z6TID/graph.json","events_json":"https://pith.science/api/pith-number/TWMJCLWO6SBRQFP6KVSN3Z6TID/events.json","paper":"https://pith.science/paper/TWMJCLWO"},"agent_actions":{"view_html":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID","download_json":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID.json","view_paper":"https://pith.science/paper/TWMJCLWO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.10372&json=true","fetch_graph":"https://pith.science/api/pith-number/TWMJCLWO6SBRQFP6KVSN3Z6TID/graph.json","fetch_events":"https://pith.science/api/pith-number/TWMJCLWO6SBRQFP6KVSN3Z6TID/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID/action/storage_attestation","attest_author":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID/action/author_attestation","sign_citation":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID/action/citation_signature","submit_replication":"https://pith.science/pith/TWMJCLWO6SBRQFP6KVSN3Z6TID/action/replication_record"}},"created_at":"2026-07-05T12:11:08.681806+00:00","updated_at":"2026-07-05T12:11:08.681806+00:00"}