{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:3O3DC37R6ZQ42SMCWFUNAHWXXT","short_pith_number":"pith:3O3DC37R","canonical_record":{"source":{"id":"2406.18485","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-26T16:51:28Z","cross_cats_sorted":[],"title_canon_sha256":"d3d3463d43f095dadfb19287508f5eed2a5d58f2dc513621ef8f96d63a2eb748","abstract_canon_sha256":"5cdb8b465fe8963c2dc751b304f13829255deebdc11e4641e13fbb5ed67eb7b2"},"schema_version":"1.0"},"canonical_sha256":"dbb6316ff1f661cd4982b168d01ed7bccfb02282b446c43e14ccb7c5f913da0a","source":{"kind":"arxiv","id":"2406.18485","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18485","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18485v1","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18485","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_12","alias_value":"3O3DC37R6ZQ4","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_16","alias_value":"3O3DC37R6ZQ42SMC","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_8","alias_value":"3O3DC37R","created_at":"2026-07-05T08:37:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:3O3DC37R6ZQ42SMCWFUNAHWXXT","target":"record","payload":{"canonical_record":{"source":{"id":"2406.18485","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-26T16:51:28Z","cross_cats_sorted":[],"title_canon_sha256":"d3d3463d43f095dadfb19287508f5eed2a5d58f2dc513621ef8f96d63a2eb748","abstract_canon_sha256":"5cdb8b465fe8963c2dc751b304f13829255deebdc11e4641e13fbb5ed67eb7b2"},"schema_version":"1.0"},"canonical_sha256":"dbb6316ff1f661cd4982b168d01ed7bccfb02282b446c43e14ccb7c5f913da0a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:37:09.910811Z","signature_b64":"aVbcgobGkfFS49Hij5+0SDhY2d+Nm+IptPgrZvvLjZixt9yXcrVqTlF3kPkuCrX6kvHrZdt8t4YfIuZzA0jJBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dbb6316ff1f661cd4982b168d01ed7bccfb02282b446c43e14ccb7c5f913da0a","last_reissued_at":"2026-07-05T08:37:09.910337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:37:09.910337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.18485","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":"2026-07-05T08:37:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nPEpR5R9KsyJ2EXVLDfBl0n2KnwySMxAUfg5Ij9qfms0VKMnnwNK8rFexlmQvRIvHgXrkumU2zHQrJEfIlfPCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T04:42:42.857171Z"},"content_sha256":"0f2019cbb61822ebab721604245a0d36d43346347c0b95f3ceaae2af80951c36","schema_version":"1.0","event_id":"sha256:0f2019cbb61822ebab721604245a0d36d43346347c0b95f3ceaae2af80951c36"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:3O3DC37R6ZQ42SMCWFUNAHWXXT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LoongTrain: Efficient Training of Long-Sequence LLMs with Head-Context Parallelism","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Diandian Gu, Guoteng Wang, Jiarui Fang, Peng Sun, Qiaoling Chen, Qinghao Hu, Shangchun Zhao, Tianwei Zhang, Ting Huang, Xin Jin, Xuanzhe Liu, Xun Chen, Yingtong Xiong, Yonggang Wen","submitted_at":"2024-06-26T16:51:28Z","abstract_excerpt":"Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or efficiency issues. We propose LoongTrain, a novel system to efficiently train LLMs with long sequences at scale. The core of LoongTrain is the 2D-Attention mechanism, which combines both head-parallel and context-parallel techniques to break the scalability constraints while maintaining efficiency. We introduce Double-Ring-Attention and analyze the performanc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18485","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/2406.18485/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":"2026-07-05T08:37:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fLbKN3ZJmfBDekvarF2kvtWBusM8/cDSBvFM/0Z5a9eTGEsOa9CAvhNIDKDpIRn9AxlytAfeN/Bt00KIQlFUCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T04:42:42.857657Z"},"content_sha256":"bb7be245b11e9efa2a960338c07eaf8c22e70a5a92e3f3c34714683f88b1c556","schema_version":"1.0","event_id":"sha256:bb7be245b11e9efa2a960338c07eaf8c22e70a5a92e3f3c34714683f88b1c556"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/bundle.json","state_url":"https://pith.science/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/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":"2026-08-19T04:42:42Z","links":{"resolver":"https://pith.science/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT","bundle":"https://pith.science/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/bundle.json","state":"https://pith.science/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/3O3DC37R6ZQ42SMCWFUNAHWXXT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:3O3DC37R6ZQ42SMCWFUNAHWXXT","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":"5cdb8b465fe8963c2dc751b304f13829255deebdc11e4641e13fbb5ed67eb7b2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-26T16:51:28Z","title_canon_sha256":"d3d3463d43f095dadfb19287508f5eed2a5d58f2dc513621ef8f96d63a2eb748"},"schema_version":"1.0","source":{"id":"2406.18485","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18485","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18485v1","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18485","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_12","alias_value":"3O3DC37R6ZQ4","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_16","alias_value":"3O3DC37R6ZQ42SMC","created_at":"2026-07-05T08:37:09Z"},{"alias_kind":"pith_short_8","alias_value":"3O3DC37R","created_at":"2026-07-05T08:37:09Z"}],"graph_snapshots":[{"event_id":"sha256:bb7be245b11e9efa2a960338c07eaf8c22e70a5a92e3f3c34714683f88b1c556","target":"graph","created_at":"2026-07-05T08:37:09Z","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/2406.18485/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Efficiently training LLMs with long sequences is important yet challenged by the massive computation and memory requirements. Sequence parallelism has been proposed to tackle these problems, but existing methods suffer from scalability or efficiency issues. We propose LoongTrain, a novel system to efficiently train LLMs with long sequences at scale. The core of LoongTrain is the 2D-Attention mechanism, which combines both head-parallel and context-parallel techniques to break the scalability constraints while maintaining efficiency. We introduce Double-Ring-Attention and analyze the performanc","authors_text":"Diandian Gu, Guoteng Wang, Jiarui Fang, Peng Sun, Qiaoling Chen, Qinghao Hu, Shangchun Zhao, Tianwei Zhang, Ting Huang, Xin Jin, Xuanzhe Liu, Xun Chen, Yingtong Xiong, Yonggang Wen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-26T16:51:28Z","title":"LoongTrain: Efficient Training of Long-Sequence LLMs with Head-Context Parallelism"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18485","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":"sha256:0f2019cbb61822ebab721604245a0d36d43346347c0b95f3ceaae2af80951c36","target":"record","created_at":"2026-07-05T08:37:09Z","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":"5cdb8b465fe8963c2dc751b304f13829255deebdc11e4641e13fbb5ed67eb7b2","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2024-06-26T16:51:28Z","title_canon_sha256":"d3d3463d43f095dadfb19287508f5eed2a5d58f2dc513621ef8f96d63a2eb748"},"schema_version":"1.0","source":{"id":"2406.18485","kind":"arxiv","version":1}},"canonical_sha256":"dbb6316ff1f661cd4982b168d01ed7bccfb02282b446c43e14ccb7c5f913da0a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"dbb6316ff1f661cd4982b168d01ed7bccfb02282b446c43e14ccb7c5f913da0a","first_computed_at":"2026-07-05T08:37:09.910337Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:37:09.910337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aVbcgobGkfFS49Hij5+0SDhY2d+Nm+IptPgrZvvLjZixt9yXcrVqTlF3kPkuCrX6kvHrZdt8t4YfIuZzA0jJBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:37:09.910811Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.18485","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0f2019cbb61822ebab721604245a0d36d43346347c0b95f3ceaae2af80951c36","sha256:bb7be245b11e9efa2a960338c07eaf8c22e70a5a92e3f3c34714683f88b1c556"],"state_sha256":"23455033f215d915622af7e4c1a774cb0afde4f777dd6cd4de950439e927db89"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rcGKAYXQeHlN2ov15rZj95FHdZyy4hD1QKPYWMJpta0gL/FZtTL7z3PMF8u4A7wLI2oTGAJTd1yffDCUX36UDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T04:42:42.860242Z","bundle_sha256":"f187d81d009d025b8c8f5eae76a8f820c4eb76f2d3e1109324456685ebc2aaf0"}}