{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:SCKWSDFRL4P5H7GZ526FVX2FSN","short_pith_number":"pith:SCKWSDFR","canonical_record":{"source":{"id":"2603.07770","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-03-08T19:20:25Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"de3ee42c336be9f1c0388fd8927ce1400360eaf62d0d6c4152fddcb89492d115","abstract_canon_sha256":"02a41d23caec35c01cb1b6d0217851f0a3d956ddab321608b8f99480cd219915"},"schema_version":"1.0"},"canonical_sha256":"9095690cb15f1fd3fcd9eebc5adf45935b7a196c3290054c5161cc8d9ae373b2","source":{"kind":"arxiv","id":"2603.07770","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.07770","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"arxiv_version","alias_value":"2603.07770v2","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.07770","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"pith_short_12","alias_value":"SCKWSDFRL4P5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"SCKWSDFRL4P5H7GZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"SCKWSDFR","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:SCKWSDFRL4P5H7GZ526FVX2FSN","target":"record","payload":{"canonical_record":{"source":{"id":"2603.07770","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-03-08T19:20:25Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"de3ee42c336be9f1c0388fd8927ce1400360eaf62d0d6c4152fddcb89492d115","abstract_canon_sha256":"02a41d23caec35c01cb1b6d0217851f0a3d956ddab321608b8f99480cd219915"},"schema_version":"1.0"},"canonical_sha256":"9095690cb15f1fd3fcd9eebc5adf45935b7a196c3290054c5161cc8d9ae373b2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:30.859872Z","signature_b64":"ppMtAdYokdb3FzvdrEicGgTJWBnl47AQGMMuPLwInflDOUMfYLmKl6CRqHdsUQ5HHfe4q0Z0wjlQqG2ESviKBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9095690cb15f1fd3fcd9eebc5adf45935b7a196c3290054c5161cc8d9ae373b2","last_reissued_at":"2026-05-18T02:44:30.859438Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:30.859438Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.07770","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t1HOEiKhUeRdjbSXYdGHVpVEEX69p/vzqn+WCHysFjHgbhpi5AJ62Ji6s5wD7TcKjffqer2STaUBPeNBUvxQAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T19:06:16.265753Z"},"content_sha256":"4ad48acc2128f2f2cb5e427cda35b14d2a0ad6269f65d82bf57d009dda62e406","schema_version":"1.0","event_id":"sha256:4ad48acc2128f2f2cb5e427cda35b14d2a0ad6269f65d82bf57d009dda62e406"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:SCKWSDFRL4P5H7GZ526FVX2FSN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ArcLight: A Lightweight LLM Inference Architecture for Many-Core CPUs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead.","cross_cats":["cs.CL"],"primary_cat":"cs.DC","authors_text":"Wanxiang Che, Xu Han, Yuxuan Li, Yuzhuang Xu","submitted_at":"2026-03-08T19:20:25Z","abstract_excerpt":"Although existing frameworks for large language model (LLM) inference on CPUs are mature, they fail to fully exploit the computation potential of many-core CPU platforms. Many-core CPUs are widely deployed in web servers and high-end networking devices, and are typically organized into multiple NUMA nodes that group cores and memory. Current frameworks largely overlook the substantial overhead of cross-NUMA memory access, limiting inference scalability and intelligence enabling on such platforms. To address this limitation, we build ArcLight, a lightweight LLM inference architecture designed f"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experimental results show that ArcLight significantly surpasses the performance ceiling of mainstream frameworks, achieving up to 46% higher inference throughput.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The assumption that cross-NUMA memory access is the dominant scalability bottleneck on many-core CPUs and that the proposed memory management and tensor parallelism will generalize without major tuning across different CPU models and workloads.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ArcLight delivers up to 46% higher LLM inference throughput on many-core CPUs by addressing cross-NUMA memory access costs through efficient memory management, thread scheduling, and tensor parallelism.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a45589c49072c820dee5bf3a5a6d6a8863a59fa7555febe75c5c2a46a0589e54"},"source":{"id":"2603.07770","kind":"arxiv","version":2},"verdict":{"id":"41e0a5b9-8921-4190-95c5-8a8b9fc6ca18","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T14:38:44.706082Z","strongest_claim":"Experimental results show that ArcLight significantly surpasses the performance ceiling of mainstream frameworks, achieving up to 46% higher inference throughput.","one_line_summary":"ArcLight delivers up to 46% higher LLM inference throughput on many-core CPUs by addressing cross-NUMA memory access costs through efficient memory management, thread scheduling, and tensor parallelism.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The assumption that cross-NUMA memory access is the dominant scalability bottleneck on many-core CPUs and that the proposed memory management and tensor parallelism will generalize without major tuning across different CPU models and workloads.","pith_extraction_headline":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"dcfd7ceea22ed4b42af8d2b5e1dbb8cf7d79f41e3e3c8c441c60e8ca9c2e77ca"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"41e0a5b9-8921-4190-95c5-8a8b9fc6ca18"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T02:44:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1I0U5OKANnF1z4IAdSv0k4iUHauw8kHrWtM9q9TY8LWv23W6c1rF5eGTCjRvkZp/FrLOMnkAW6aMz1MGbFGVDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-28T19:06:16.266547Z"},"content_sha256":"e2fcb129c5ffdf1967fd92013c35644aeb8f0c2e64886ce7fb5f341b3554b55a","schema_version":"1.0","event_id":"sha256:e2fcb129c5ffdf1967fd92013c35644aeb8f0c2e64886ce7fb5f341b3554b55a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/bundle.json","state_url":"https://pith.science/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-28T19:06:16Z","links":{"resolver":"https://pith.science/pith/SCKWSDFRL4P5H7GZ526FVX2FSN","bundle":"https://pith.science/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/bundle.json","state":"https://pith.science/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SCKWSDFRL4P5H7GZ526FVX2FSN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SCKWSDFRL4P5H7GZ526FVX2FSN","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"02a41d23caec35c01cb1b6d0217851f0a3d956ddab321608b8f99480cd219915","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-03-08T19:20:25Z","title_canon_sha256":"de3ee42c336be9f1c0388fd8927ce1400360eaf62d0d6c4152fddcb89492d115"},"schema_version":"1.0","source":{"id":"2603.07770","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.07770","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"arxiv_version","alias_value":"2603.07770v2","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.07770","created_at":"2026-05-18T02:44:30Z"},{"alias_kind":"pith_short_12","alias_value":"SCKWSDFRL4P5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"SCKWSDFRL4P5H7GZ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"SCKWSDFR","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:e2fcb129c5ffdf1967fd92013c35644aeb8f0c2e64886ce7fb5f341b3554b55a","target":"graph","created_at":"2026-05-18T02:44:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experimental results show that ArcLight significantly surpasses the performance ceiling of mainstream frameworks, achieving up to 46% higher inference throughput."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The assumption that cross-NUMA memory access is the dominant scalability bottleneck on many-core CPUs and that the proposed memory management and tensor parallelism will generalize without major tuning across different CPU models and workloads."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ArcLight delivers up to 46% higher LLM inference throughput on many-core CPUs by addressing cross-NUMA memory access costs through efficient memory management, thread scheduling, and tensor parallelism."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead."}],"snapshot_sha256":"a45589c49072c820dee5bf3a5a6d6a8863a59fa7555febe75c5c2a46a0589e54"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"dcfd7ceea22ed4b42af8d2b5e1dbb8cf7d79f41e3e3c8c441c60e8ca9c2e77ca"},"paper":{"abstract_excerpt":"Although existing frameworks for large language model (LLM) inference on CPUs are mature, they fail to fully exploit the computation potential of many-core CPU platforms. Many-core CPUs are widely deployed in web servers and high-end networking devices, and are typically organized into multiple NUMA nodes that group cores and memory. Current frameworks largely overlook the substantial overhead of cross-NUMA memory access, limiting inference scalability and intelligence enabling on such platforms. To address this limitation, we build ArcLight, a lightweight LLM inference architecture designed f","authors_text":"Wanxiang Che, Xu Han, Yuxuan Li, Yuzhuang Xu","cross_cats":["cs.CL"],"headline":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-03-08T19:20:25Z","title":"ArcLight: A Lightweight LLM Inference Architecture for Many-Core CPUs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.07770","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T14:38:44.706082Z","id":"41e0a5b9-8921-4190-95c5-8a8b9fc6ca18","model_set":{"reader":"grok-4.3"},"one_line_summary":"ArcLight delivers up to 46% higher LLM inference throughput on many-core CPUs by addressing cross-NUMA memory access costs through efficient memory management, thread scheduling, and tensor parallelism.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"ArcLight achieves up to 46% higher LLM inference throughput on many-core CPUs by reducing cross-NUMA memory access overhead.","strongest_claim":"Experimental results show that ArcLight significantly surpasses the performance ceiling of mainstream frameworks, achieving up to 46% higher inference throughput.","weakest_assumption":"The assumption that cross-NUMA memory access is the dominant scalability bottleneck on many-core CPUs and that the proposed memory management and tensor parallelism will generalize without major tuning across different CPU models and workloads."}},"verdict_id":"41e0a5b9-8921-4190-95c5-8a8b9fc6ca18"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4ad48acc2128f2f2cb5e427cda35b14d2a0ad6269f65d82bf57d009dda62e406","target":"record","created_at":"2026-05-18T02:44:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"02a41d23caec35c01cb1b6d0217851f0a3d956ddab321608b8f99480cd219915","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-03-08T19:20:25Z","title_canon_sha256":"de3ee42c336be9f1c0388fd8927ce1400360eaf62d0d6c4152fddcb89492d115"},"schema_version":"1.0","source":{"id":"2603.07770","kind":"arxiv","version":2}},"canonical_sha256":"9095690cb15f1fd3fcd9eebc5adf45935b7a196c3290054c5161cc8d9ae373b2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9095690cb15f1fd3fcd9eebc5adf45935b7a196c3290054c5161cc8d9ae373b2","first_computed_at":"2026-05-18T02:44:30.859438Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:30.859438Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ppMtAdYokdb3FzvdrEicGgTJWBnl47AQGMMuPLwInflDOUMfYLmKl6CRqHdsUQ5HHfe4q0Z0wjlQqG2ESviKBQ==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:30.859872Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.07770","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ad48acc2128f2f2cb5e427cda35b14d2a0ad6269f65d82bf57d009dda62e406","sha256:e2fcb129c5ffdf1967fd92013c35644aeb8f0c2e64886ce7fb5f341b3554b55a"],"state_sha256":"62f62df3336861da0b9b59eb8fbd7d0b85c9309e83357e45238284685051da18"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mCra43Y6olmdf8fssUGEJFixf5ttiXkBleTHILO6D8OCa0Q+rbu7KztnrNDHCGXudG/6JfQSoDdvD35V6QWxBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-28T19:06:16.270488Z","bundle_sha256":"02d26767e5b42cb67773ef0c8a80f5c93fc585a875d5d7c762a8ade5fc25a58e"}}