{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:B7IGJXJFFBDII7PJO3FPHKSC44","short_pith_number":"pith:B7IGJXJF","canonical_record":{"source":{"id":"2508.08457","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-11T20:30:31Z","cross_cats_sorted":["cs.ET"],"title_canon_sha256":"a0bbbd61ffe6ec28200c7c6e70f73bee7ebf0bb04d689566d02620489d826118","abstract_canon_sha256":"9d62d1cece5e3cbcedf1cf462629aa9c509b295d06564dfc72a7cc3b335648e1"},"schema_version":"1.0"},"canonical_sha256":"0fd064dd252846847de976caf3aa42e71e8469372da6f8e84e9d0ffe968b5beb","source":{"kind":"arxiv","id":"2508.08457","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.08457","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"arxiv_version","alias_value":"2508.08457v1","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.08457","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_12","alias_value":"B7IGJXJFFBDI","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_16","alias_value":"B7IGJXJFFBDII7PJ","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_8","alias_value":"B7IGJXJF","created_at":"2026-07-05T11:52:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:B7IGJXJFFBDII7PJO3FPHKSC44","target":"record","payload":{"canonical_record":{"source":{"id":"2508.08457","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-11T20:30:31Z","cross_cats_sorted":["cs.ET"],"title_canon_sha256":"a0bbbd61ffe6ec28200c7c6e70f73bee7ebf0bb04d689566d02620489d826118","abstract_canon_sha256":"9d62d1cece5e3cbcedf1cf462629aa9c509b295d06564dfc72a7cc3b335648e1"},"schema_version":"1.0"},"canonical_sha256":"0fd064dd252846847de976caf3aa42e71e8469372da6f8e84e9d0ffe968b5beb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:52:28.811524Z","signature_b64":"eTcTcQdHrgW6/fu5KNzNfpOkLP7baETevTNo9oH0jr4SpOT1ubKiA+NWGfIP2DpTv6k6+7UwYtZ4LDLFxKgVBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0fd064dd252846847de976caf3aa42e71e8469372da6f8e84e9d0ffe968b5beb","last_reissued_at":"2026-07-05T11:52:28.811029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:52:28.811029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.08457","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:52:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"YTlm3SGWfMkiBf0GW3OBkojyXbTzeao8iQlAANejwogEvyUXkfgvYxgzK+mqB9MyGrvJwRpMwZapXREA/ECfDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:18:14.041547Z"},"content_sha256":"0a94a73053870829e8e15d79be3a1ba244cb0a0a223d60245e5291ad1cc49adb","schema_version":"1.0","event_id":"sha256:0a94a73053870829e8e15d79be3a1ba244cb0a0a223d60245e5291ad1cc49adb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:B7IGJXJFFBDII7PJO3FPHKSC44","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Architecting Long-Context LLM Acceleration with Packing-Prefetch Scheduler and Ultra-Large Capacity On-Chip Memories","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.ET"],"primary_cat":"cs.AR","authors_text":"Faaiq Waqar, Hanchen Yang, Harsono Simka, Ming-Yen Lee, Muhammed Ahosan Ul Karim, Shimeng Yu","submitted_at":"2025-08-11T20:30:31Z","abstract_excerpt":"Long-context Large Language Model (LLM) inference faces increasing compute bottlenecks as attention calculations scale with context length, primarily due to the growing KV-cache transfer overhead that saturates High Bandwidth Memory (HBM). While prefetching techniques mitigate cache misses by fetching KV data in advance, their spatial and temporal benefits present new opportunities to exploit. This work proposes a packing-prefetch scheduling architecture with monolithic 3D (M3D) back-end-of-line (BEOL) compatible embedded memories with ultra-large on-chip capacity to accelerate long-context LL"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.08457","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/2508.08457/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:52:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rk2OrrUM+PGc3Q3/kDnSil/PtTFUv95A+1D2wIaoUu7IhFwkLNTfzP9LlCLWFDsqKxwVAnFUo8G5hoDSVxNcAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:18:14.042111Z"},"content_sha256":"92f17b01dbc468e8dac2e98fe6af5328f1bd6726290df50ca1fb9d4d532db0cd","schema_version":"1.0","event_id":"sha256:92f17b01dbc468e8dac2e98fe6af5328f1bd6726290df50ca1fb9d4d532db0cd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B7IGJXJFFBDII7PJO3FPHKSC44/bundle.json","state_url":"https://pith.science/pith/B7IGJXJFFBDII7PJO3FPHKSC44/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B7IGJXJFFBDII7PJO3FPHKSC44/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":"2026-08-09T13:18:14Z","links":{"resolver":"https://pith.science/pith/B7IGJXJFFBDII7PJO3FPHKSC44","bundle":"https://pith.science/pith/B7IGJXJFFBDII7PJO3FPHKSC44/bundle.json","state":"https://pith.science/pith/B7IGJXJFFBDII7PJO3FPHKSC44/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B7IGJXJFFBDII7PJO3FPHKSC44/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:B7IGJXJFFBDII7PJO3FPHKSC44","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":"9d62d1cece5e3cbcedf1cf462629aa9c509b295d06564dfc72a7cc3b335648e1","cross_cats_sorted":["cs.ET"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-11T20:30:31Z","title_canon_sha256":"a0bbbd61ffe6ec28200c7c6e70f73bee7ebf0bb04d689566d02620489d826118"},"schema_version":"1.0","source":{"id":"2508.08457","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.08457","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"arxiv_version","alias_value":"2508.08457v1","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.08457","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_12","alias_value":"B7IGJXJFFBDI","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_16","alias_value":"B7IGJXJFFBDII7PJ","created_at":"2026-07-05T11:52:28Z"},{"alias_kind":"pith_short_8","alias_value":"B7IGJXJF","created_at":"2026-07-05T11:52:28Z"}],"graph_snapshots":[{"event_id":"sha256:92f17b01dbc468e8dac2e98fe6af5328f1bd6726290df50ca1fb9d4d532db0cd","target":"graph","created_at":"2026-07-05T11:52:28Z","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/2508.08457/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Long-context Large Language Model (LLM) inference faces increasing compute bottlenecks as attention calculations scale with context length, primarily due to the growing KV-cache transfer overhead that saturates High Bandwidth Memory (HBM). While prefetching techniques mitigate cache misses by fetching KV data in advance, their spatial and temporal benefits present new opportunities to exploit. This work proposes a packing-prefetch scheduling architecture with monolithic 3D (M3D) back-end-of-line (BEOL) compatible embedded memories with ultra-large on-chip capacity to accelerate long-context LL","authors_text":"Faaiq Waqar, Hanchen Yang, Harsono Simka, Ming-Yen Lee, Muhammed Ahosan Ul Karim, Shimeng Yu","cross_cats":["cs.ET"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-11T20:30:31Z","title":"Architecting Long-Context LLM Acceleration with Packing-Prefetch Scheduler and Ultra-Large Capacity On-Chip Memories"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.08457","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":"sha256:0a94a73053870829e8e15d79be3a1ba244cb0a0a223d60245e5291ad1cc49adb","target":"record","created_at":"2026-07-05T11:52:28Z","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":"9d62d1cece5e3cbcedf1cf462629aa9c509b295d06564dfc72a7cc3b335648e1","cross_cats_sorted":["cs.ET"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AR","submitted_at":"2025-08-11T20:30:31Z","title_canon_sha256":"a0bbbd61ffe6ec28200c7c6e70f73bee7ebf0bb04d689566d02620489d826118"},"schema_version":"1.0","source":{"id":"2508.08457","kind":"arxiv","version":1}},"canonical_sha256":"0fd064dd252846847de976caf3aa42e71e8469372da6f8e84e9d0ffe968b5beb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0fd064dd252846847de976caf3aa42e71e8469372da6f8e84e9d0ffe968b5beb","first_computed_at":"2026-07-05T11:52:28.811029Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:52:28.811029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"eTcTcQdHrgW6/fu5KNzNfpOkLP7baETevTNo9oH0jr4SpOT1ubKiA+NWGfIP2DpTv6k6+7UwYtZ4LDLFxKgVBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:52:28.811524Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.08457","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0a94a73053870829e8e15d79be3a1ba244cb0a0a223d60245e5291ad1cc49adb","sha256:92f17b01dbc468e8dac2e98fe6af5328f1bd6726290df50ca1fb9d4d532db0cd"],"state_sha256":"e8cc55cc1d6bd18f3936439ec04768b41e804241693eacb18bf1795fa3a8c9e5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uUnT4eAkmFUCbjQ0iXv2HCaeOvL7A+AO1xjv9ac7uwvf1UAfUli/0zyYYrEws0/0E9lAzjF6VdxXizqEfJ03DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T13:18:14.045648Z","bundle_sha256":"aacd0f896a1526d4b5106cf9387557b30034cf4c93fde31f42e32941e7ffb63b"}}