{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:B5WA6BGSARH6457R4HMIGAPXDT","short_pith_number":"pith:B5WA6BGS","canonical_record":{"source":{"id":"2401.06761","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-12T18:50:36Z","cross_cats_sorted":[],"title_canon_sha256":"8aec5782dba780a17e96af1265ed7696b15254a740212c987119825ebd87ab6b","abstract_canon_sha256":"69d0a5cb5e012e36206bec1da2e1e610660f93da07711599d030071bb1c39451"},"schema_version":"1.0"},"canonical_sha256":"0f6c0f04d2044fee77f1e1d88301f71cc78898af96082ac86e1773d3b42991d4","source":{"kind":"arxiv","id":"2401.06761","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.06761","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"arxiv_version","alias_value":"2401.06761v1","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.06761","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_12","alias_value":"B5WA6BGSARH6","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_16","alias_value":"B5WA6BGSARH6457R","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_8","alias_value":"B5WA6BGS","created_at":"2026-07-05T07:33:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:B5WA6BGSARH6457R4HMIGAPXDT","target":"record","payload":{"canonical_record":{"source":{"id":"2401.06761","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-12T18:50:36Z","cross_cats_sorted":[],"title_canon_sha256":"8aec5782dba780a17e96af1265ed7696b15254a740212c987119825ebd87ab6b","abstract_canon_sha256":"69d0a5cb5e012e36206bec1da2e1e610660f93da07711599d030071bb1c39451"},"schema_version":"1.0"},"canonical_sha256":"0f6c0f04d2044fee77f1e1d88301f71cc78898af96082ac86e1773d3b42991d4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:33:04.939978Z","signature_b64":"IH55/qruSIbluOYzvr090Kkr5tRCAd1ZWUiQWcaJFOzA9g9Dtq89c3Nw841MZ6qgrxv3FkK9zkSwlQybfUpaDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f6c0f04d2044fee77f1e1d88301f71cc78898af96082ac86e1773d3b42991d4","last_reissued_at":"2026-07-05T07:33:04.939565Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:33:04.939565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.06761","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":"2026-07-05T07:33:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RWxqqhjleeTJX7M65TjHHgUalM39q4IYQ8nAuFg9RbHh9a2IWRQ0rqRKTeaBv6DeFAG4lDx6H9giPfldqaN5Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:50:04.791267Z"},"content_sha256":"aaab1b863238010df17721e4a5bfbc087ea3eff9c9f5fefcb42569fbc68f44c5","schema_version":"1.0","event_id":"sha256:aaab1b863238010df17721e4a5bfbc087ea3eff9c9f5fefcb42569fbc68f44c5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:B5WA6BGSARH6457R4HMIGAPXDT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Aohan Zeng, Bowen Wang, Jie Tang, Mingdao Liu, Peng Zhang, Yuxiao Dong","submitted_at":"2024-01-12T18:50:36Z","abstract_excerpt":"The massive adoption of large language models (LLMs) demands efficient deployment strategies. However, the auto-regressive decoding process, which is fundamental to how most LLMs generate text, poses challenges to achieve efficient serving. In this work, we introduce a parallel auto-regressive generation method. By instruct-tuning on general domain data that contains hierarchical structures, we enable LLMs to independently plan their generation process and perform auto-parallel auto-regressive (APAR) generation, significantly reducing the number of generation steps. APAR alone can achieve up t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.06761","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/2401.06761/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":"2026-07-05T07:33:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"edF2ULrn8DlOEdp62AtA17LcXggiYuMJ+XlHq7rSdd34HpEbsFF6nz3EvXCNKku2O75ctIM0Ag/QT6ju+ZKbBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:50:04.791959Z"},"content_sha256":"4f40d3729388a35e787734c8fe492fcaeaccc0a9de3b273c7b262c964da6d618","schema_version":"1.0","event_id":"sha256:4f40d3729388a35e787734c8fe492fcaeaccc0a9de3b273c7b262c964da6d618"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B5WA6BGSARH6457R4HMIGAPXDT/bundle.json","state_url":"https://pith.science/pith/B5WA6BGSARH6457R4HMIGAPXDT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B5WA6BGSARH6457R4HMIGAPXDT/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":"2026-08-04T11:50:04Z","links":{"resolver":"https://pith.science/pith/B5WA6BGSARH6457R4HMIGAPXDT","bundle":"https://pith.science/pith/B5WA6BGSARH6457R4HMIGAPXDT/bundle.json","state":"https://pith.science/pith/B5WA6BGSARH6457R4HMIGAPXDT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B5WA6BGSARH6457R4HMIGAPXDT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:B5WA6BGSARH6457R4HMIGAPXDT","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":"69d0a5cb5e012e36206bec1da2e1e610660f93da07711599d030071bb1c39451","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-12T18:50:36Z","title_canon_sha256":"8aec5782dba780a17e96af1265ed7696b15254a740212c987119825ebd87ab6b"},"schema_version":"1.0","source":{"id":"2401.06761","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.06761","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"arxiv_version","alias_value":"2401.06761v1","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.06761","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_12","alias_value":"B5WA6BGSARH6","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_16","alias_value":"B5WA6BGSARH6457R","created_at":"2026-07-05T07:33:04Z"},{"alias_kind":"pith_short_8","alias_value":"B5WA6BGS","created_at":"2026-07-05T07:33:04Z"}],"graph_snapshots":[{"event_id":"sha256:4f40d3729388a35e787734c8fe492fcaeaccc0a9de3b273c7b262c964da6d618","target":"graph","created_at":"2026-07-05T07:33:04Z","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/2401.06761/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The massive adoption of large language models (LLMs) demands efficient deployment strategies. However, the auto-regressive decoding process, which is fundamental to how most LLMs generate text, poses challenges to achieve efficient serving. In this work, we introduce a parallel auto-regressive generation method. By instruct-tuning on general domain data that contains hierarchical structures, we enable LLMs to independently plan their generation process and perform auto-parallel auto-regressive (APAR) generation, significantly reducing the number of generation steps. APAR alone can achieve up t","authors_text":"Aohan Zeng, Bowen Wang, Jie Tang, Mingdao Liu, Peng Zhang, Yuxiao Dong","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-12T18:50:36Z","title":"APAR: LLMs Can Do Auto-Parallel Auto-Regressive Decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.06761","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:aaab1b863238010df17721e4a5bfbc087ea3eff9c9f5fefcb42569fbc68f44c5","target":"record","created_at":"2026-07-05T07:33:04Z","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":"69d0a5cb5e012e36206bec1da2e1e610660f93da07711599d030071bb1c39451","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-01-12T18:50:36Z","title_canon_sha256":"8aec5782dba780a17e96af1265ed7696b15254a740212c987119825ebd87ab6b"},"schema_version":"1.0","source":{"id":"2401.06761","kind":"arxiv","version":1}},"canonical_sha256":"0f6c0f04d2044fee77f1e1d88301f71cc78898af96082ac86e1773d3b42991d4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0f6c0f04d2044fee77f1e1d88301f71cc78898af96082ac86e1773d3b42991d4","first_computed_at":"2026-07-05T07:33:04.939565Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:33:04.939565Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"IH55/qruSIbluOYzvr090Kkr5tRCAd1ZWUiQWcaJFOzA9g9Dtq89c3Nw841MZ6qgrxv3FkK9zkSwlQybfUpaDA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:33:04.939978Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.06761","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:aaab1b863238010df17721e4a5bfbc087ea3eff9c9f5fefcb42569fbc68f44c5","sha256:4f40d3729388a35e787734c8fe492fcaeaccc0a9de3b273c7b262c964da6d618"],"state_sha256":"42d06995e265e8dc8e2600d8cc2487ec9b81ead296764e9299437d7c1409cc44"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0+EgwEsFT6vn7adMULqZFQJUfBlzlc5Zexu00S3BR7Oyi34CrZQd2nFBG/QZz92mIkZrZAJc+hbFdANyM66vAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T11:50:04.797811Z","bundle_sha256":"ab14751a3c2a290f99e6431542186b690016dd7fcfe3c3394549ddb22da81843"}}