{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FOITHG2CC7JZL7GPRLPZAGSSBI","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":"dc97e2bc597d231ea146f45ef6a719ca4afcd4d5f76c4cfd3c7e289b51c5d4d7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-03T20:20:33Z","title_canon_sha256":"0cdb5d0aa08dbce6a3dfea05b0ab6bc968332c35963bd21839fb2baf18648880"},"schema_version":"1.0","source":{"id":"2507.03153","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.03153","created_at":"2026-07-05T11:31:53Z"},{"alias_kind":"arxiv_version","alias_value":"2507.03153v1","created_at":"2026-07-05T11:31:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.03153","created_at":"2026-07-05T11:31:53Z"},{"alias_kind":"pith_short_12","alias_value":"FOITHG2CC7JZ","created_at":"2026-07-05T11:31:53Z"},{"alias_kind":"pith_short_16","alias_value":"FOITHG2CC7JZL7GP","created_at":"2026-07-05T11:31:53Z"},{"alias_kind":"pith_short_8","alias_value":"FOITHG2C","created_at":"2026-07-05T11:31:53Z"}],"graph_snapshots":[{"event_id":"sha256:9af7ec60fe44acb27f78f52fc57114661846a45c2ac6be7ab3610a26ab5f2728","target":"graph","created_at":"2026-07-05T11:31:53Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2507.03153/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Scaling inference for large language models (LLMs) is increasingly constrained by limited GPU memory, especially due to growing key-value (KV) caches required for long-context generation. While existing approaches offload KV caches to CPU memory or apply sparse attention to reduce GPU load, they often underutilize CPU compute resources and compromise accuracy. We present HGCA, a hybrid CPU-GPU attention mechanism that enables scalable, high-throughput LLM inference with near-full attention quality. HGCA performs dense attention on recently generated KV entries retained in GPU memory and parall","authors_text":"Chen Zhong, Hui Lu, Jia Rao, Lingfeng Xiang, Peiran Du, Song Jiang, Weishu Deng, Yujie Yang, Zhen Lin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-03T20:20:33Z","title":"HGCA: Hybrid GPU-CPU Attention for Long Context LLM Inference"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.03153","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:2fd75a87b824021716655c7d94922b0802c2b4a0b13d1ded48c2378baa2a072f","target":"record","created_at":"2026-07-05T11:31:53Z","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":"dc97e2bc597d231ea146f45ef6a719ca4afcd4d5f76c4cfd3c7e289b51c5d4d7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-07-03T20:20:33Z","title_canon_sha256":"0cdb5d0aa08dbce6a3dfea05b0ab6bc968332c35963bd21839fb2baf18648880"},"schema_version":"1.0","source":{"id":"2507.03153","kind":"arxiv","version":1}},"canonical_sha256":"2b91339b4217d395fccf8adf901a520a2c82cb82f2c6d336f7a3abb19f46a970","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2b91339b4217d395fccf8adf901a520a2c82cb82f2c6d336f7a3abb19f46a970","first_computed_at":"2026-07-05T11:31:53.559799Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:31:53.559799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lYwGSVYHkDAyKUKFxF8cJny9uM7oBQY9/vrhC7Ek5/ztZOYER5zMwOk0MWUhhKJcsswzHJy3N8MT39/iwhV9AA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:31:53.560264Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.03153","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2fd75a87b824021716655c7d94922b0802c2b4a0b13d1ded48c2378baa2a072f","sha256:9af7ec60fe44acb27f78f52fc57114661846a45c2ac6be7ab3610a26ab5f2728"],"state_sha256":"a3fc50f42f1270bcad7159472a56eca2f500a7b47362e7f76904a6184b0eb5c7"}