{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:HH2MM2EJGG5IOL3PCPN6EV4EHO","short_pith_number":"pith:HH2MM2EJ","canonical_record":{"source":{"id":"2405.14009","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-05-22T21:35:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8d1514e53fc378a3450a96900d148f889ba15327aabf28cb4ffd7106c3d68f84","abstract_canon_sha256":"931284f09d73c7e05c249c684c991456a1edd28657bf774134759e0bdfa71bad"},"schema_version":"1.0"},"canonical_sha256":"39f4c6688931ba872f6f13dbe257843bb1552dc160b86de035a192e75e21d357","source":{"kind":"arxiv","id":"2405.14009","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.14009","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2405.14009v2","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14009","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"HH2MM2EJGG5I","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"HH2MM2EJGG5IOL3P","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"HH2MM2EJ","created_at":"2026-07-05T09:11:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:HH2MM2EJGG5IOL3PCPN6EV4EHO","target":"record","payload":{"canonical_record":{"source":{"id":"2405.14009","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-05-22T21:35:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"8d1514e53fc378a3450a96900d148f889ba15327aabf28cb4ffd7106c3d68f84","abstract_canon_sha256":"931284f09d73c7e05c249c684c991456a1edd28657bf774134759e0bdfa71bad"},"schema_version":"1.0"},"canonical_sha256":"39f4c6688931ba872f6f13dbe257843bb1552dc160b86de035a192e75e21d357","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:11:47.336197Z","signature_b64":"YVWJNna3vU9sDBMXI13hgDdHfxIwjS6rlyk1yRg5pKop9L8Zo6mpQ7fQoxRQwPHPkveIx3iipw8yP9MoER7ODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"39f4c6688931ba872f6f13dbe257843bb1552dc160b86de035a192e75e21d357","last_reissued_at":"2026-07-05T09:11:47.335664Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:11:47.335664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.14009","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kr26ffxZvIjDugjMXlWL5ItS0Stoz7/NJT/5sEmWrIp3ILE46D4Bi/L4yScKK723tb3SiuxXcABy9r9B8SyEAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T04:05:07.891340Z"},"content_sha256":"599fef6aca8f7f3655d884eecc6cf0e5e7a41b39e47b7a4b8cf257444573567b","schema_version":"1.0","event_id":"sha256:599fef6aca8f7f3655d884eecc6cf0e5e7a41b39e47b7a4b8cf257444573567b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:HH2MM2EJGG5IOL3PCPN6EV4EHO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.DC","authors_text":"Athinagoras Skiadopoulos, Christos Kozyrakis, Mark Zhao, Swapnil Gandhi","submitted_at":"2024-05-22T21:35:56Z","abstract_excerpt":"Training large Deep Neural Network (DNN) models requires thousands of GPUs over the course of several days or weeks. At this scale, failures are frequent and can have a big impact on training throughput. Utilizing spare GPU servers to mitigate performance loss becomes increasingly costly as model sizes grow. ReCycle is a system designed for efficient DNN training in the presence of failures, without relying on spare servers. It exploits the inherent functional redundancy in distributed training systems -- where servers across data-parallel groups store the same model parameters -- and pipeline"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14009","kind":"arxiv","version":2},"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/2405.14009/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":"2026-07-05T09:11:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eYjvUQjtR03DJ42m+ZaTXmZYKnfyG9gVTMVy7HmGq5W0F0VNLzr6Wl6SQQ3fkR1E/Li+tNb7ZZkldUVD/q5GBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T04:05:07.891719Z"},"content_sha256":"a53f1a9fd270346ab86dbd4960992c18fccfed52039692a429b97b2d8f7c3f16","schema_version":"1.0","event_id":"sha256:a53f1a9fd270346ab86dbd4960992c18fccfed52039692a429b97b2d8f7c3f16"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/bundle.json","state_url":"https://pith.science/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/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":"2026-07-09T04:05:07Z","links":{"resolver":"https://pith.science/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO","bundle":"https://pith.science/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/bundle.json","state":"https://pith.science/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HH2MM2EJGG5IOL3PCPN6EV4EHO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:HH2MM2EJGG5IOL3PCPN6EV4EHO","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":"931284f09d73c7e05c249c684c991456a1edd28657bf774134759e0bdfa71bad","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-05-22T21:35:56Z","title_canon_sha256":"8d1514e53fc378a3450a96900d148f889ba15327aabf28cb4ffd7106c3d68f84"},"schema_version":"1.0","source":{"id":"2405.14009","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.14009","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"arxiv_version","alias_value":"2405.14009v2","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.14009","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_12","alias_value":"HH2MM2EJGG5I","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_16","alias_value":"HH2MM2EJGG5IOL3P","created_at":"2026-07-05T09:11:47Z"},{"alias_kind":"pith_short_8","alias_value":"HH2MM2EJ","created_at":"2026-07-05T09:11:47Z"}],"graph_snapshots":[{"event_id":"sha256:a53f1a9fd270346ab86dbd4960992c18fccfed52039692a429b97b2d8f7c3f16","target":"graph","created_at":"2026-07-05T09:11:47Z","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/2405.14009/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training large Deep Neural Network (DNN) models requires thousands of GPUs over the course of several days or weeks. At this scale, failures are frequent and can have a big impact on training throughput. Utilizing spare GPU servers to mitigate performance loss becomes increasingly costly as model sizes grow. ReCycle is a system designed for efficient DNN training in the presence of failures, without relying on spare servers. It exploits the inherent functional redundancy in distributed training systems -- where servers across data-parallel groups store the same model parameters -- and pipeline","authors_text":"Athinagoras Skiadopoulos, Christos Kozyrakis, Mark Zhao, Swapnil Gandhi","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-05-22T21:35:56Z","title":"ReCycle: Resilient Training of Large DNNs using Pipeline Adaptation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.14009","kind":"arxiv","version":2},"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":"sha256:599fef6aca8f7f3655d884eecc6cf0e5e7a41b39e47b7a4b8cf257444573567b","target":"record","created_at":"2026-07-05T09:11:47Z","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":"931284f09d73c7e05c249c684c991456a1edd28657bf774134759e0bdfa71bad","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-05-22T21:35:56Z","title_canon_sha256":"8d1514e53fc378a3450a96900d148f889ba15327aabf28cb4ffd7106c3d68f84"},"schema_version":"1.0","source":{"id":"2405.14009","kind":"arxiv","version":2}},"canonical_sha256":"39f4c6688931ba872f6f13dbe257843bb1552dc160b86de035a192e75e21d357","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"39f4c6688931ba872f6f13dbe257843bb1552dc160b86de035a192e75e21d357","first_computed_at":"2026-07-05T09:11:47.335664Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:11:47.335664Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YVWJNna3vU9sDBMXI13hgDdHfxIwjS6rlyk1yRg5pKop9L8Zo6mpQ7fQoxRQwPHPkveIx3iipw8yP9MoER7ODg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:11:47.336197Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.14009","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:599fef6aca8f7f3655d884eecc6cf0e5e7a41b39e47b7a4b8cf257444573567b","sha256:a53f1a9fd270346ab86dbd4960992c18fccfed52039692a429b97b2d8f7c3f16"],"state_sha256":"a8c8bdf7f748360ec0749d678c210e4b4578858c718ebc8644cd1c1a1b5151b5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ocg1SxdeK0niCidULk5UeCUXui5HJQRNsxfkMTCCq5KXMsOutNVr8x8vRj3JdRonn1p6MYqhNrpteP27LPldBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-09T04:05:07.894014Z","bundle_sha256":"dda6d2bbdeaab0a6c17b8afbb47812e1a6c8fce392e7ea7c359ef0771be1355c"}}