{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:UBC674LNXM3WO3RFPQYDAQ6PXA","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":"f9490a5f5052389c2756ffdf0f62fb5e6fb19ace01eacac48670cd6098683ee2","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2026-08-03T03:43:14Z","title_canon_sha256":"ce46f9a5bc9059583d4150660bec1b2170a4d3bc21138e68310dac660dc39ce9"},"schema_version":"1.0","source":{"id":"2608.01651","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.01651","created_at":"2026-08-04T02:06:20Z"},{"alias_kind":"arxiv_version","alias_value":"2608.01651v1","created_at":"2026-08-04T02:06:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.01651","created_at":"2026-08-04T02:06:20Z"},{"alias_kind":"pith_short_12","alias_value":"UBC674LNXM3W","created_at":"2026-08-04T02:06:20Z"},{"alias_kind":"pith_short_16","alias_value":"UBC674LNXM3WO3RF","created_at":"2026-08-04T02:06:20Z"},{"alias_kind":"pith_short_8","alias_value":"UBC674LN","created_at":"2026-08-04T02:06:20Z"}],"graph_snapshots":[{"event_id":"sha256:be1cbcbf330d061582d28fe306fae9d39ae121fe427ca825a85b7a1b1d64b3d9","target":"graph","created_at":"2026-08-04T02:06:20Z","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/2608.01651/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Hybrid-attention large language models combine full attention with recurrent linear attention to reduce long-context inference costs, yet their autoregressive decoding remains memory-bound. Tree speculative decoding offers an attractive acceleration path, but existing tree-speculation systems are designed around the key--value caches of full-attention models. On hybrid models, they traverse recurrent layers branch by branch and materialize a full state for every proposal node, causing verification latency and transient memory to scale poorly with tree and batch sizes. We present Bole, a kernel","authors_text":"Chengying Huan, Chen Tian, Chiran You, Fangxin Liu, Jiajun Zheng, Jie Zhang, Juelu Zhang, Li Wang, Xiabao Wu, Yi Su, Yongchao Liu, Zhan Qiu","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2026-08-03T03:43:14Z","title":"Bole: Efficient Tree Speculation for Hybrid-Attention Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01651","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":"sha256:79cde1e3ce5258120f4ab5067d0b17e6f1833a1ac3f84a41d829cb49c32971e4","target":"record","created_at":"2026-08-04T02:06:20Z","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":"f9490a5f5052389c2756ffdf0f62fb5e6fb19ace01eacac48670cd6098683ee2","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2026-08-03T03:43:14Z","title_canon_sha256":"ce46f9a5bc9059583d4150660bec1b2170a4d3bc21138e68310dac660dc39ce9"},"schema_version":"1.0","source":{"id":"2608.01651","kind":"arxiv","version":1}},"canonical_sha256":"a045eff16dbb37676e257c303043cfb80b0c89d9096df5d02da49625a7eadaa2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a045eff16dbb37676e257c303043cfb80b0c89d9096df5d02da49625a7eadaa2","first_computed_at":"2026-08-04T02:06:20.381374Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T02:06:20.381374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MHawhyauUT+gdjPAVX5T//EwvAmJwGjKiPgVTXK8TtmaXyys4GylgFrTFlQBfzJOspcgH0ggQWl7X84RVC5jAw==","signature_status":"signed_v1","signed_at":"2026-08-04T02:06:20.382905Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.01651","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79cde1e3ce5258120f4ab5067d0b17e6f1833a1ac3f84a41d829cb49c32971e4","sha256:be1cbcbf330d061582d28fe306fae9d39ae121fe427ca825a85b7a1b1d64b3d9"],"state_sha256":"7c5b509316fa4ce290b6068747f5fbdab6e801b3c46e10ba1e1e01c41b82ddad"}