{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:AEEUJMWLLE7ETX7KR5G2LBL2VW","short_pith_number":"pith:AEEUJMWL","canonical_record":{"source":{"id":"2504.06214","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-08T16:58:58Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"f6478b48c5b18849486f5c4cba41bf4287c8287ade5fc251cd145cafb287c364","abstract_canon_sha256":"aa61b93a97d1d8f8b134d23039ab90f19414790e00cd43e17ae141687e8b902c"},"schema_version":"1.0"},"canonical_sha256":"010944b2cb593e49dfea8f4da5857aad9611b76c3b47b4a0d4cf67f3d28ac6f8","source":{"kind":"arxiv","id":"2504.06214","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.06214","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"arxiv_version","alias_value":"2504.06214v1","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06214","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_12","alias_value":"AEEUJMWLLE7E","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_16","alias_value":"AEEUJMWLLE7ETX7K","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_8","alias_value":"AEEUJMWL","created_at":"2026-07-05T10:46:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:AEEUJMWLLE7ETX7KR5G2LBL2VW","target":"record","payload":{"canonical_record":{"source":{"id":"2504.06214","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-08T16:58:58Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"f6478b48c5b18849486f5c4cba41bf4287c8287ade5fc251cd145cafb287c364","abstract_canon_sha256":"aa61b93a97d1d8f8b134d23039ab90f19414790e00cd43e17ae141687e8b902c"},"schema_version":"1.0"},"canonical_sha256":"010944b2cb593e49dfea8f4da5857aad9611b76c3b47b4a0d4cf67f3d28ac6f8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:17.411781Z","signature_b64":"1/I9JoG/wRcSMoMqaR66BiChKWju+UJLFo8BEHgDnTdgt0lyH+ap4afy5+yUW6q79EKCepSV+dOMrtXHyIGiBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"010944b2cb593e49dfea8f4da5857aad9611b76c3b47b4a0d4cf67f3d28ac6f8","last_reissued_at":"2026-07-05T10:46:17.411278Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:17.411278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.06214","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":"2026-07-05T10:46:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HzRiuCD5oZFDM8Nkdkhs3f/VDLdWHeHRvSZ1BF9Dk1O+ybDBJr2lgDFsQ7yAGbh43Po6kxs+UZ8xSpknVVOKAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T03:02:25.897574Z"},"content_sha256":"7d7a70e3dc865e3c7cfbcb972c770f7f8a3d3b7bcbba8d508e9e1e165467d901","schema_version":"1.0","event_id":"sha256:7d7a70e3dc865e3c7cfbcb972c770f7f8a3d3b7bcbba8d508e9e1e165467d901"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:AEEUJMWLLE7ETX7KR5G2LBL2VW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Bo Li, Boxin Wang, Bryan Catanzaro, Chejian Xu, Mohammad Shoeybi, Peng Xu, Wei Ping, Zihan Liu","submitted_at":"2025-04-08T16:58:58Z","abstract_excerpt":"Long-context capabilities are essential for a wide range of applications, including document and video understanding, in-context learning, and inference-time scaling, all of which require models to process and reason over long sequences of text and multimodal data. In this work, we introduce a efficient training recipe for building ultra-long context LLMs from aligned instruct model, pushing the boundaries of context lengths from 128K to 1M, 2M, and 4M tokens. Our approach leverages efficient continued pretraining strategies to extend the context window and employs effective instruction tuning"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06214","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/2504.06214/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":"2026-07-05T10:46:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uhSkN/B4LESYtWYpBngflBZBdvxFhCAI+0vd1BvaXILpYMU9fj9JNhbgv9ftzxJ7G10/0E0G/0DiJiwRWehuCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T03:02:25.898160Z"},"content_sha256":"742bbeaa97608ac485799b34414b64b6e30b55bb01d6ac51c305a90455f8c8fb","schema_version":"1.0","event_id":"sha256:742bbeaa97608ac485799b34414b64b6e30b55bb01d6ac51c305a90455f8c8fb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/bundle.json","state_url":"https://pith.science/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/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":"2026-08-09T03:02:25Z","links":{"resolver":"https://pith.science/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW","bundle":"https://pith.science/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/bundle.json","state":"https://pith.science/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AEEUJMWLLE7ETX7KR5G2LBL2VW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:AEEUJMWLLE7ETX7KR5G2LBL2VW","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":"aa61b93a97d1d8f8b134d23039ab90f19414790e00cd43e17ae141687e8b902c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-08T16:58:58Z","title_canon_sha256":"f6478b48c5b18849486f5c4cba41bf4287c8287ade5fc251cd145cafb287c364"},"schema_version":"1.0","source":{"id":"2504.06214","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.06214","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"arxiv_version","alias_value":"2504.06214v1","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.06214","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_12","alias_value":"AEEUJMWLLE7E","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_16","alias_value":"AEEUJMWLLE7ETX7K","created_at":"2026-07-05T10:46:17Z"},{"alias_kind":"pith_short_8","alias_value":"AEEUJMWL","created_at":"2026-07-05T10:46:17Z"}],"graph_snapshots":[{"event_id":"sha256:742bbeaa97608ac485799b34414b64b6e30b55bb01d6ac51c305a90455f8c8fb","target":"graph","created_at":"2026-07-05T10:46:17Z","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/2504.06214/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Long-context capabilities are essential for a wide range of applications, including document and video understanding, in-context learning, and inference-time scaling, all of which require models to process and reason over long sequences of text and multimodal data. In this work, we introduce a efficient training recipe for building ultra-long context LLMs from aligned instruct model, pushing the boundaries of context lengths from 128K to 1M, 2M, and 4M tokens. Our approach leverages efficient continued pretraining strategies to extend the context window and employs effective instruction tuning","authors_text":"Bo Li, Boxin Wang, Bryan Catanzaro, Chejian Xu, Mohammad Shoeybi, Peng Xu, Wei Ping, Zihan Liu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-08T16:58:58Z","title":"From 128K to 4M: Efficient Training of Ultra-Long Context Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.06214","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":"sha256:7d7a70e3dc865e3c7cfbcb972c770f7f8a3d3b7bcbba8d508e9e1e165467d901","target":"record","created_at":"2026-07-05T10:46:17Z","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":"aa61b93a97d1d8f8b134d23039ab90f19414790e00cd43e17ae141687e8b902c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-08T16:58:58Z","title_canon_sha256":"f6478b48c5b18849486f5c4cba41bf4287c8287ade5fc251cd145cafb287c364"},"schema_version":"1.0","source":{"id":"2504.06214","kind":"arxiv","version":1}},"canonical_sha256":"010944b2cb593e49dfea8f4da5857aad9611b76c3b47b4a0d4cf67f3d28ac6f8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"010944b2cb593e49dfea8f4da5857aad9611b76c3b47b4a0d4cf67f3d28ac6f8","first_computed_at":"2026-07-05T10:46:17.411278Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:46:17.411278Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1/I9JoG/wRcSMoMqaR66BiChKWju+UJLFo8BEHgDnTdgt0lyH+ap4afy5+yUW6q79EKCepSV+dOMrtXHyIGiBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:46:17.411781Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.06214","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7d7a70e3dc865e3c7cfbcb972c770f7f8a3d3b7bcbba8d508e9e1e165467d901","sha256:742bbeaa97608ac485799b34414b64b6e30b55bb01d6ac51c305a90455f8c8fb"],"state_sha256":"11ef89cf32d708e5f0faab3e5ab6d4688185292c3636fb45b068037f47c4e5eb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eZUCIqV8kQvdRPEDxXV5bIni1QPG77sEEtRZ8HvXgrSQRpZNwJDUU0rVRfR8Wjfz5Srwas499mA1E2iNzzlUBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T03:02:25.902497Z","bundle_sha256":"e0cee335dc51e7376a8f6ffda0937df8fdc9d77d2bfedae0d8bc2724e07de6ef"}}