{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:UBC674LNXM3WO3RFPQYDAQ6PXA","short_pith_number":"pith:UBC674LN","canonical_record":{"source":{"id":"2608.01651","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2026-08-03T03:43:14Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"ce46f9a5bc9059583d4150660bec1b2170a4d3bc21138e68310dac660dc39ce9","abstract_canon_sha256":"f9490a5f5052389c2756ffdf0f62fb5e6fb19ace01eacac48670cd6098683ee2"},"schema_version":"1.0"},"canonical_sha256":"a045eff16dbb37676e257c303043cfb80b0c89d9096df5d02da49625a7eadaa2","source":{"kind":"arxiv","id":"2608.01651","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"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:UBC674LNXM3WO3RFPQYDAQ6PXA","target":"record","payload":{"canonical_record":{"source":{"id":"2608.01651","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.DC","submitted_at":"2026-08-03T03:43:14Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"ce46f9a5bc9059583d4150660bec1b2170a4d3bc21138e68310dac660dc39ce9","abstract_canon_sha256":"f9490a5f5052389c2756ffdf0f62fb5e6fb19ace01eacac48670cd6098683ee2"},"schema_version":"1.0"},"canonical_sha256":"a045eff16dbb37676e257c303043cfb80b0c89d9096df5d02da49625a7eadaa2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:06:20.382905Z","signature_b64":"MHawhyauUT+gdjPAVX5T//EwvAmJwGjKiPgVTXK8TtmaXyys4GylgFrTFlQBfzJOspcgH0ggQWl7X84RVC5jAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a045eff16dbb37676e257c303043cfb80b0c89d9096df5d02da49625a7eadaa2","last_reissued_at":"2026-08-04T02:06:20.381374Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:06:20.381374Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.01651","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":"2026-08-04T02:06:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9OKmfnAVoXDxDPDXG4HHwM7ZiyIoIe7bfk+kX1DVfDsEX1jNLiTYF/HK4XCwbPNctzusgLl3YozaCoqpHj4HDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T18:30:54.683225Z"},"content_sha256":"79cde1e3ce5258120f4ab5067d0b17e6f1833a1ac3f84a41d829cb49c32971e4","schema_version":"1.0","event_id":"sha256:79cde1e3ce5258120f4ab5067d0b17e6f1833a1ac3f84a41d829cb49c32971e4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:UBC674LNXM3WO3RFPQYDAQ6PXA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Bole: Efficient Tree Speculation for Hybrid-Attention Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.DC","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","submitted_at":"2026-08-03T03:43:14Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01651","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/2608.01651/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":"2026-08-04T02:06:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mDYYeJMt4tBloF8p1jX1t/TrVyu+fglOuSf1E7GsbxlbwnMaLMJcBuLlW6I2CdM0NZjSnXjn2528Zr7WatJsBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T18:30:54.683687Z"},"content_sha256":"be1cbcbf330d061582d28fe306fae9d39ae121fe427ca825a85b7a1b1d64b3d9","schema_version":"1.0","event_id":"sha256:be1cbcbf330d061582d28fe306fae9d39ae121fe427ca825a85b7a1b1d64b3d9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/bundle.json","state_url":"https://pith.science/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/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":"2026-08-05T18:30:54Z","links":{"resolver":"https://pith.science/pith/UBC674LNXM3WO3RFPQYDAQ6PXA","bundle":"https://pith.science/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/bundle.json","state":"https://pith.science/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UBC674LNXM3WO3RFPQYDAQ6PXA/bundle.json"},"state":{"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"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"72uGgfnu7rRc7wgVACIJQhHMD0CqfIZMRcS1dNiBQZW85wppbfPdOI4HIRYhkXncvQ7y3LeXqiimTMvWR72LBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T18:30:54.687801Z","bundle_sha256":"8f1ce0af7625ada6f55d48171a3b812dcd95dd9048583db87d5069d1e9f0da40"}}