{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:5EBCKYJR64SMOIFCHSXDKEYHMR","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":"133c1b84fcaf5eacf75464d9cea0f67f5afdd29eb70163758f6ad35470236570","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-07T18:53:01Z","title_canon_sha256":"190f585c1cd56f3371d17acc2a588f19f019e0f1d17ff5dffa4f1fde19a9a1da"},"schema_version":"1.0","source":{"id":"2402.05099","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.05099","created_at":"2026-07-05T08:18:05Z"},{"alias_kind":"arxiv_version","alias_value":"2402.05099v2","created_at":"2026-07-05T08:18:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.05099","created_at":"2026-07-05T08:18:05Z"},{"alias_kind":"pith_short_12","alias_value":"5EBCKYJR64SM","created_at":"2026-07-05T08:18:05Z"},{"alias_kind":"pith_short_16","alias_value":"5EBCKYJR64SMOIFC","created_at":"2026-07-05T08:18:05Z"},{"alias_kind":"pith_short_8","alias_value":"5EBCKYJR","created_at":"2026-07-05T08:18:05Z"}],"graph_snapshots":[{"event_id":"sha256:059adc1ef3ded98b039786fefeb9e2a6d4ce13cb99c9e93c506953d5367927f3","target":"graph","created_at":"2026-07-05T08:18:05Z","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/2402.05099/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer-based large language models (LLMs) are now deployed to hundreds of millions of users. LLM inference is commonly performed on batches of sequences that share a prefix, such as few-shot examples or a chatbot system prompt. Decoding in this large-batch setting can be bottlenecked by the attention operation, which reads large key-value (KV) caches from memory and computes inefficient matrix-vector products for every sequence in the batch. In this work, we introduce Hydragen, a hardware-aware exact implementation of attention with shared prefixes. Hydragen computes attention over the sh","authors_text":"Azalia Mirhoseini, Bradley Brown, Christopher R\\'e, Daniel Y. Fu, Jordan Juravsky, Ryan Ehrlich","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-07T18:53:01Z","title":"Hydragen: High-Throughput LLM Inference with Shared Prefixes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.05099","kind":"arxiv","version":2},"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:f19adb422fca6d1dc3d5ee46051f11d902a9929b7c96c5810a7997fd418e0bce","target":"record","created_at":"2026-07-05T08:18:05Z","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":"133c1b84fcaf5eacf75464d9cea0f67f5afdd29eb70163758f6ad35470236570","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-02-07T18:53:01Z","title_canon_sha256":"190f585c1cd56f3371d17acc2a588f19f019e0f1d17ff5dffa4f1fde19a9a1da"},"schema_version":"1.0","source":{"id":"2402.05099","kind":"arxiv","version":2}},"canonical_sha256":"e902256131f724c720a23cae351307645a4d96898e4aed251630e1cd3a06bf1b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e902256131f724c720a23cae351307645a4d96898e4aed251630e1cd3a06bf1b","first_computed_at":"2026-07-05T08:18:05.052571Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:18:05.052571Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WCG3HnF9Ct7NkWthbNUxRkfI325bgxCLrWGP95d8aYCvl+FL0DFLeGB25TMWY3wyAZ/cPwgaajkXblOSdIuGCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:18:05.053055Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.05099","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f19adb422fca6d1dc3d5ee46051f11d902a9929b7c96c5810a7997fd418e0bce","sha256:059adc1ef3ded98b039786fefeb9e2a6d4ce13cb99c9e93c506953d5367927f3"],"state_sha256":"fe7ac81d0e90ddc01979e7485d4cc16ef1a3d6bdff69a5ee19cd0bda6065e7a9"}