{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4UNKEC2NSJTB2ETIARQUS6FT6I","short_pith_number":"pith:4UNKEC2N","canonical_record":{"source":{"id":"2505.16710","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T14:11:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3c6a9f6fa415d5d7835235af60af8827a49633421a350beb49e788db7d48bfe6","abstract_canon_sha256":"45d18069056acc2045d3cfcef5724c8cda5a9cae5b4c9f955a2eb262c447277e"},"schema_version":"1.0"},"canonical_sha256":"e51aa20b4d92661d126804614978b3f2231215a667bff4bcb5b80c6db8ce38df","source":{"kind":"arxiv","id":"2505.16710","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.16710","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"arxiv_version","alias_value":"2505.16710v1","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16710","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_12","alias_value":"4UNKEC2NSJTB","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_16","alias_value":"4UNKEC2NSJTB2ETI","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_8","alias_value":"4UNKEC2N","created_at":"2026-07-05T11:07:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4UNKEC2NSJTB2ETIARQUS6FT6I","target":"record","payload":{"canonical_record":{"source":{"id":"2505.16710","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T14:11:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3c6a9f6fa415d5d7835235af60af8827a49633421a350beb49e788db7d48bfe6","abstract_canon_sha256":"45d18069056acc2045d3cfcef5724c8cda5a9cae5b4c9f955a2eb262c447277e"},"schema_version":"1.0"},"canonical_sha256":"e51aa20b4d92661d126804614978b3f2231215a667bff4bcb5b80c6db8ce38df","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:39.610185Z","signature_b64":"4TuMNOkCbwcIKcvDIFqxPly0s3BVc7MEqxjOXwkOoZM3OpNiB3s7HOX6Dovdxjtg+Z0U4lgPmsUfEcOsuGmqCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e51aa20b4d92661d126804614978b3f2231215a667bff4bcb5b80c6db8ce38df","last_reissued_at":"2026-07-05T11:07:39.609676Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:39.609676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.16710","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":"2026-07-05T11:07:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ibzHIkR7eu5M4Fel1/4ptkgEY6gqvLSFjypzUMZUFB+4bsVnDXJzJcWfN5lFRry/nCGG9UijY0N2EODnIHrFCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T02:45:36.322783Z"},"content_sha256":"655cca393167f55b4674b3c9ddabdd0ad0beb384f0b0a8adfcac86951ffe10b2","schema_version":"1.0","event_id":"sha256:655cca393167f55b4674b3c9ddabdd0ad0beb384f0b0a8adfcac86951ffe10b2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4UNKEC2NSJTB2ETIARQUS6FT6I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training Long-Context LLMs Efficiently via Chunk-wise Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Daohai Yu, Gen Luo, Rongrong Ji, Wenhao Li, Yuxin Zhang","submitted_at":"2025-05-22T14:11:34Z","abstract_excerpt":"While long-context large language models (LLMs) exhibit remarkable document processing capabilities, their prohibitively high training costs often hinder customized applications. To mitigate this issue, we propose \\textit{Sequential Chunk-wise Optimization} (SeCO), a memory-efficient training paradigm that partitions lengthy inputs into manageable chunks. Each chunk independently constructs its computational graph and performs localized backpropagation, ensuring that only one chunk's forward activations are stored in memory. Building on SeCO, we further introduce \\textit{Sparse Chunk-wise Opti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16710","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/2505.16710/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":"2026-07-05T11:07:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Mw9Amffw/3ZDR1NDbXDoBZCULsF8+5axKAlC2qYWQDmatBZNzuaK2M5rau94JKXj4LOmgPTpOplGl9GfAEepBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T02:45:36.323743Z"},"content_sha256":"897db37c2cb1350c28d9e5d17e87050ea49224af5e09129e57da26d50823c96a","schema_version":"1.0","event_id":"sha256:897db37c2cb1350c28d9e5d17e87050ea49224af5e09129e57da26d50823c96a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/bundle.json","state_url":"https://pith.science/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/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":"2026-08-04T02:45:36Z","links":{"resolver":"https://pith.science/pith/4UNKEC2NSJTB2ETIARQUS6FT6I","bundle":"https://pith.science/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/bundle.json","state":"https://pith.science/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4UNKEC2NSJTB2ETIARQUS6FT6I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4UNKEC2NSJTB2ETIARQUS6FT6I","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":"45d18069056acc2045d3cfcef5724c8cda5a9cae5b4c9f955a2eb262c447277e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T14:11:34Z","title_canon_sha256":"3c6a9f6fa415d5d7835235af60af8827a49633421a350beb49e788db7d48bfe6"},"schema_version":"1.0","source":{"id":"2505.16710","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.16710","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"arxiv_version","alias_value":"2505.16710v1","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.16710","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_12","alias_value":"4UNKEC2NSJTB","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_16","alias_value":"4UNKEC2NSJTB2ETI","created_at":"2026-07-05T11:07:39Z"},{"alias_kind":"pith_short_8","alias_value":"4UNKEC2N","created_at":"2026-07-05T11:07:39Z"}],"graph_snapshots":[{"event_id":"sha256:897db37c2cb1350c28d9e5d17e87050ea49224af5e09129e57da26d50823c96a","target":"graph","created_at":"2026-07-05T11:07:39Z","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/2505.16710/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While long-context large language models (LLMs) exhibit remarkable document processing capabilities, their prohibitively high training costs often hinder customized applications. To mitigate this issue, we propose \\textit{Sequential Chunk-wise Optimization} (SeCO), a memory-efficient training paradigm that partitions lengthy inputs into manageable chunks. Each chunk independently constructs its computational graph and performs localized backpropagation, ensuring that only one chunk's forward activations are stored in memory. Building on SeCO, we further introduce \\textit{Sparse Chunk-wise Opti","authors_text":"Daohai Yu, Gen Luo, Rongrong Ji, Wenhao Li, Yuxin Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T14:11:34Z","title":"Training Long-Context LLMs Efficiently via Chunk-wise Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.16710","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":"sha256:655cca393167f55b4674b3c9ddabdd0ad0beb384f0b0a8adfcac86951ffe10b2","target":"record","created_at":"2026-07-05T11:07:39Z","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":"45d18069056acc2045d3cfcef5724c8cda5a9cae5b4c9f955a2eb262c447277e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-22T14:11:34Z","title_canon_sha256":"3c6a9f6fa415d5d7835235af60af8827a49633421a350beb49e788db7d48bfe6"},"schema_version":"1.0","source":{"id":"2505.16710","kind":"arxiv","version":1}},"canonical_sha256":"e51aa20b4d92661d126804614978b3f2231215a667bff4bcb5b80c6db8ce38df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e51aa20b4d92661d126804614978b3f2231215a667bff4bcb5b80c6db8ce38df","first_computed_at":"2026-07-05T11:07:39.609676Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:07:39.609676Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4TuMNOkCbwcIKcvDIFqxPly0s3BVc7MEqxjOXwkOoZM3OpNiB3s7HOX6Dovdxjtg+Z0U4lgPmsUfEcOsuGmqCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:07:39.610185Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.16710","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:655cca393167f55b4674b3c9ddabdd0ad0beb384f0b0a8adfcac86951ffe10b2","sha256:897db37c2cb1350c28d9e5d17e87050ea49224af5e09129e57da26d50823c96a"],"state_sha256":"4a803c5ade1ea22b7cd4193286e6b5ff3750bd650228bf5d38894c0832842282"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rlo1chH0zzyIA9OyHo2dBuwafnnLnH+vS4pFvL62v0dB2umY2LcqtT0vqK6iMcqqzHN+oKdEDy6WWleUhuOlBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T02:45:36.339178Z","bundle_sha256":"1278e288d6e0f5b5899b674d629d11d0db890f758ee052a13f0c988b9aa6e435"}}