{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:M2I6JEJKFHSL36BSOT2P66LF7U","short_pith_number":"pith:M2I6JEJK","schema_version":"1.0","canonical_sha256":"6691e4912a29e4bdf83274f4ff7965fd3eef034466d51ceaaee1904275a20524","source":{"kind":"arxiv","id":"2411.17309","version":1},"attestation_state":"computed","paper":{"title":"PIM-AI: A Novel Architecture for High-Efficiency LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DC","cs.ET"],"primary_cat":"cs.AR","authors_text":"Cristobal Ortega, Renaud Ayrignac, Yann Falevoz","submitted_at":"2024-11-26T10:54:19Z","abstract_excerpt":"Large Language Models (LLMs) have become essential in a variety of applications due to their advanced language understanding and generation capabilities. However, their computational and memory requirements pose significant challenges to traditional hardware architectures. Processing-in-Memory (PIM), which integrates computational units directly into memory chips, offers several advantages for LLM inference, including reduced data transfer bottlenecks and improved power efficiency.\n  This paper introduces PIM-AI, a novel DDR5/LPDDR5 PIM architecture designed for LLM inference without modifying"},"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":"2411.17309","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2024-11-26T10:54:19Z","cross_cats_sorted":["cs.AI","cs.DC","cs.ET"],"title_canon_sha256":"eb91b85e7fba8293b95714b0dda4c8608794d3b48fd979b232ced740a08e6263","abstract_canon_sha256":"ad14f1dc28b30a9dd2b3f77f7051e715565256b48cae5d0bca6531f695f9b249"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:24.338784Z","signature_b64":"9Jv5vrOOr+tqcBdAMUQA64hwt5Yb1ycAvF4pnqY0klQ6RVrPjVBslWNjT3+mBY7w+EiYFiUR8xXumF1d4XCaAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6691e4912a29e4bdf83274f4ff7965fd3eef034466d51ceaaee1904275a20524","last_reissued_at":"2026-07-05T09:41:24.338290Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:24.338290Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PIM-AI: A Novel Architecture for High-Efficiency LLM Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DC","cs.ET"],"primary_cat":"cs.AR","authors_text":"Cristobal Ortega, Renaud Ayrignac, Yann Falevoz","submitted_at":"2024-11-26T10:54:19Z","abstract_excerpt":"Large Language Models (LLMs) have become essential in a variety of applications due to their advanced language understanding and generation capabilities. However, their computational and memory requirements pose significant challenges to traditional hardware architectures. Processing-in-Memory (PIM), which integrates computational units directly into memory chips, offers several advantages for LLM inference, including reduced data transfer bottlenecks and improved power efficiency.\n  This paper introduces PIM-AI, a novel DDR5/LPDDR5 PIM architecture designed for LLM inference without modifying"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.17309","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/2411.17309/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":"2411.17309","created_at":"2026-07-05T09:41:24.338352+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.17309v1","created_at":"2026-07-05T09:41:24.338352+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.17309","created_at":"2026-07-05T09:41:24.338352+00:00"},{"alias_kind":"pith_short_12","alias_value":"M2I6JEJKFHSL","created_at":"2026-07-05T09:41:24.338352+00:00"},{"alias_kind":"pith_short_16","alias_value":"M2I6JEJKFHSL36BS","created_at":"2026-07-05T09:41:24.338352+00:00"},{"alias_kind":"pith_short_8","alias_value":"M2I6JEJK","created_at":"2026-07-05T09:41:24.338352+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.30553","citing_title":"COSM: A Cooperative Scheduling Framework for Concurrent PIM and CPU Execution on Mobile Devices","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2606.30553","citing_title":"COSM: A Cooperative Scheduling Framework for Concurrent PIM and CPU Execution on Mobile Devices","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25522","citing_title":"Co-Designing Graph-based Approximate Nearest Neighbor Search at Billion Scale for Processing-in-Memory","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25522","citing_title":"Co-Designing Graph-based Approximate Nearest Neighbor Search at Billion Scale for Processing-in-Memory","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07935","citing_title":"The Hyperscale Lottery: How State-Space Models Have Sacrificed Edge Efficiency","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U","json":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U.json","graph_json":"https://pith.science/api/pith-number/M2I6JEJKFHSL36BSOT2P66LF7U/graph.json","events_json":"https://pith.science/api/pith-number/M2I6JEJKFHSL36BSOT2P66LF7U/events.json","paper":"https://pith.science/paper/M2I6JEJK"},"agent_actions":{"view_html":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U","download_json":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U.json","view_paper":"https://pith.science/paper/M2I6JEJK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.17309&json=true","fetch_graph":"https://pith.science/api/pith-number/M2I6JEJKFHSL36BSOT2P66LF7U/graph.json","fetch_events":"https://pith.science/api/pith-number/M2I6JEJKFHSL36BSOT2P66LF7U/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U/action/storage_attestation","attest_author":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U/action/author_attestation","sign_citation":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U/action/citation_signature","submit_replication":"https://pith.science/pith/M2I6JEJKFHSL36BSOT2P66LF7U/action/replication_record"}},"created_at":"2026-07-05T09:41:24.338352+00:00","updated_at":"2026-07-05T09:41:24.338352+00:00"}