{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:QAUZV7WSQ7YPVYUSRJAUVTARQF","short_pith_number":"pith:QAUZV7WS","canonical_record":{"source":{"id":"2507.00394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-01T03:11:18Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"577ca3a6e51298ba80ef49498d15c55ee96b1011673f8c276c286089089c7dda","abstract_canon_sha256":"759672836b69aed5ceba598ab5f901e50285244a97acef7d0f0f3115c26f958a"},"schema_version":"1.0"},"canonical_sha256":"80299afed287f0fae2928a414acc1181603a02c67c083e4b988fc7489d9b7770","source":{"kind":"arxiv","id":"2507.00394","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.00394","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"arxiv_version","alias_value":"2507.00394v1","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.00394","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_12","alias_value":"QAUZV7WSQ7YP","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_16","alias_value":"QAUZV7WSQ7YPVYUS","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_8","alias_value":"QAUZV7WS","created_at":"2026-07-05T11:30:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:QAUZV7WSQ7YPVYUSRJAUVTARQF","target":"record","payload":{"canonical_record":{"source":{"id":"2507.00394","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-01T03:11:18Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"577ca3a6e51298ba80ef49498d15c55ee96b1011673f8c276c286089089c7dda","abstract_canon_sha256":"759672836b69aed5ceba598ab5f901e50285244a97acef7d0f0f3115c26f958a"},"schema_version":"1.0"},"canonical_sha256":"80299afed287f0fae2928a414acc1181603a02c67c083e4b988fc7489d9b7770","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:30:08.143952Z","signature_b64":"zHCXtxXvuuUUNyJb+4fpViBBhWNYEaJ/oh8gPRBZiWWA3Z9GUyz3wQzgGJWyGO5QXwsXWsccDAjYJy/wBmweDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80299afed287f0fae2928a414acc1181603a02c67c083e4b988fc7489d9b7770","last_reissued_at":"2026-07-05T11:30:08.143492Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:30:08.143492Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2507.00394","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":"2026-07-05T11:30:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"QYYvltWLM3ep7sM0alfTdiRvWf8DdwN5s4v0XAnLlXX1WhayE22JchAP+itEMvvp9t2knJLR0aZ7DdPdrsRGCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:29:55.296797Z"},"content_sha256":"0358de9142a8010f085e402c2ecc78f4570836c4c2aa90a5fab7a0a097b82711","schema_version":"1.0","event_id":"sha256:0358de9142a8010f085e402c2ecc78f4570836c4c2aa90a5fab7a0a097b82711"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:QAUZV7WSQ7YPVYUSRJAUVTARQF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"HelixPipe: Efficient Distributed Training of Long Sequence Transformers with Attention Parallel Pipeline Parallelism","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Geng Zhang, Shenggan Cheng, Xuanlei Zhao, Yang You, Ziming Liu","submitted_at":"2025-07-01T03:11:18Z","abstract_excerpt":"As transformer sequence lengths grow, existing pipeline parallelisms incur suboptimal performance due to the quadratic attention computation and the substantial memory overhead. To relieve these challenges, we propose HelixPipe, a novel pipeline parallelism for long sequence transformer training. First, HelixPipe introduces attention parallel partition, which schedules attention computations of different micro batches across different pipeline stages in parallel, reducing pipeline bubbles. Second, it employs a two-fold first-in-last-out micro batch schedule to balance memory usage and overlap "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.00394","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/2507.00394/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":"2026-07-05T11:30:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WV7hzIpm3WnkHCaN7u7w01U9Oq+cRT0+dszMb2Bw7uXvSNf7xr3aCgcJhKDRytMCAjEp09qKWNWMWrjDwFHnAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T18:29:55.297326Z"},"content_sha256":"7b1ea48ba81efa05286a987bdb2349ff1a345273110e32409b528ed65e967ef2","schema_version":"1.0","event_id":"sha256:7b1ea48ba81efa05286a987bdb2349ff1a345273110e32409b528ed65e967ef2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/bundle.json","state_url":"https://pith.science/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/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":"2026-08-17T18:29:55Z","links":{"resolver":"https://pith.science/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF","bundle":"https://pith.science/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/bundle.json","state":"https://pith.science/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QAUZV7WSQ7YPVYUSRJAUVTARQF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QAUZV7WSQ7YPVYUSRJAUVTARQF","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":"759672836b69aed5ceba598ab5f901e50285244a97acef7d0f0f3115c26f958a","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-01T03:11:18Z","title_canon_sha256":"577ca3a6e51298ba80ef49498d15c55ee96b1011673f8c276c286089089c7dda"},"schema_version":"1.0","source":{"id":"2507.00394","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2507.00394","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"arxiv_version","alias_value":"2507.00394v1","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.00394","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_12","alias_value":"QAUZV7WSQ7YP","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_16","alias_value":"QAUZV7WSQ7YPVYUS","created_at":"2026-07-05T11:30:08Z"},{"alias_kind":"pith_short_8","alias_value":"QAUZV7WS","created_at":"2026-07-05T11:30:08Z"}],"graph_snapshots":[{"event_id":"sha256:7b1ea48ba81efa05286a987bdb2349ff1a345273110e32409b528ed65e967ef2","target":"graph","created_at":"2026-07-05T11:30:08Z","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/2507.00394/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As transformer sequence lengths grow, existing pipeline parallelisms incur suboptimal performance due to the quadratic attention computation and the substantial memory overhead. To relieve these challenges, we propose HelixPipe, a novel pipeline parallelism for long sequence transformer training. First, HelixPipe introduces attention parallel partition, which schedules attention computations of different micro batches across different pipeline stages in parallel, reducing pipeline bubbles. Second, it employs a two-fold first-in-last-out micro batch schedule to balance memory usage and overlap ","authors_text":"Geng Zhang, Shenggan Cheng, Xuanlei Zhao, Yang You, Ziming Liu","cross_cats":["cs.DC"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-01T03:11:18Z","title":"HelixPipe: Efficient Distributed Training of Long Sequence Transformers with Attention Parallel Pipeline Parallelism"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.00394","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":"sha256:0358de9142a8010f085e402c2ecc78f4570836c4c2aa90a5fab7a0a097b82711","target":"record","created_at":"2026-07-05T11:30:08Z","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":"759672836b69aed5ceba598ab5f901e50285244a97acef7d0f0f3115c26f958a","cross_cats_sorted":["cs.DC"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-07-01T03:11:18Z","title_canon_sha256":"577ca3a6e51298ba80ef49498d15c55ee96b1011673f8c276c286089089c7dda"},"schema_version":"1.0","source":{"id":"2507.00394","kind":"arxiv","version":1}},"canonical_sha256":"80299afed287f0fae2928a414acc1181603a02c67c083e4b988fc7489d9b7770","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"80299afed287f0fae2928a414acc1181603a02c67c083e4b988fc7489d9b7770","first_computed_at":"2026-07-05T11:30:08.143492Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:30:08.143492Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zHCXtxXvuuUUNyJb+4fpViBBhWNYEaJ/oh8gPRBZiWWA3Z9GUyz3wQzgGJWyGO5QXwsXWsccDAjYJy/wBmweDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:30:08.143952Z","signed_message":"canonical_sha256_bytes"},"source_id":"2507.00394","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0358de9142a8010f085e402c2ecc78f4570836c4c2aa90a5fab7a0a097b82711","sha256:7b1ea48ba81efa05286a987bdb2349ff1a345273110e32409b528ed65e967ef2"],"state_sha256":"f9f40cb7be275b06ac38f299e29822ef6f46cc6903e7e46b0ede9b58d5441aa6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AOHO6VADxiXZSupw9BjKAw4FruVVQc26/50IsLJblD3MgEKBzkaE6ztJHXxLFlpLQif2BKQn3HHV+DEidLA4Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T18:29:55.302869Z","bundle_sha256":"6184427fad7283f87970f33039da7c2f3550aa7b99f16c768bce24d4bb310ce6"}}