{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MXSZ36Q3R5V43FCS5HBFPUOGI4","short_pith_number":"pith:MXSZ36Q3","canonical_record":{"source":{"id":"2605.17879","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T05:40:52Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"fdc0cfc9f785a4663754bbf9c239908277ba75e1c3f43a06d34ce0ec46461fe9","abstract_canon_sha256":"ebf386d7d4e2fb534079894272145c9d8a4da7614cec862474d8a7f5c6b3a962"},"schema_version":"1.0"},"canonical_sha256":"65e59dfa1b8f6bcd9452e9c257d1c6472e64efa35fc500ce7a1a36979da4fffd","source":{"kind":"arxiv","id":"2605.17879","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17879","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17879v1","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17879","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_12","alias_value":"MXSZ36Q3R5V4","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_16","alias_value":"MXSZ36Q3R5V43FCS","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_8","alias_value":"MXSZ36Q3","created_at":"2026-05-20T00:05:03Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MXSZ36Q3R5V43FCS5HBFPUOGI4","target":"record","payload":{"canonical_record":{"source":{"id":"2605.17879","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T05:40:52Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"fdc0cfc9f785a4663754bbf9c239908277ba75e1c3f43a06d34ce0ec46461fe9","abstract_canon_sha256":"ebf386d7d4e2fb534079894272145c9d8a4da7614cec862474d8a7f5c6b3a962"},"schema_version":"1.0"},"canonical_sha256":"65e59dfa1b8f6bcd9452e9c257d1c6472e64efa35fc500ce7a1a36979da4fffd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:05:03.296498Z","signature_b64":"R/OYpIaFpBlz2VfZ4HWray4MFQbFoNBTKxbJ36ZLiNkfny9vPmbC2+IZ7pFrnNFFbrFOVHsz4jX9eW+jaDQXCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"65e59dfa1b8f6bcd9452e9c257d1c6472e64efa35fc500ce7a1a36979da4fffd","last_reissued_at":"2026-05-20T00:05:03.295563Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:05:03.295563Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.17879","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":"2026-05-20T00:05:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F8sEBKhKfetLJBg/BtNRw/MevacOGY24icIAhYx/hUKPT7kkEM/SbmOhE5CFjW7n8Fd72XbtpUeak+8xfwa2Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T05:39:22.767405Z"},"content_sha256":"21c489d587b871f037a7939566a57cb32b4c28d82c2620c809696d9796577419","schema_version":"1.0","event_id":"sha256:21c489d587b871f037a7939566a57cb32b4c28d82c2620c809696d9796577419"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MXSZ36Q3R5V43FCS5HBFPUOGI4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Guard: Scalable Straggler Detection and Node Health Management for Large-Scale Training","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.DC","authors_text":"Abhinandan Patni, Alexander Zhipa, Anthony Ko, Ashvin Nihalani, Binxuan Huang, Cong Cheng, Congzhu Lin, Guanliang Liu, Jack Wittmayer, Josh Wu, Mi Sun, Parthasarathy Govindarajen, Rejith George Joseph, Rory Na, Vijay Rajakumar, Yinghong Liu, Zoe Zeng","submitted_at":"2026-05-18T05:40:52Z","abstract_excerpt":"Training frontier-scale foundation models involves coordinating tens of thousands of GPUs over multi-month runs, where even minor performance degradations can accumulate into substantial efficiency losses. Existing health-check mechanisms, such as NCCL tests or GPU burn-in, primarily focus on functional correctness and often fail to detect fail-slow behaviors that silently degrade system performance. In this paper, we present Guard, a scalable system for detecting stragglers and ensuring node health in large-scale training clusters. Guard combines lightweight online performance monitoring duri"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17879","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/2605.17879/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":"2026-05-20T00:05:03Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Zz5UXXiuXcxrO3lMbvsOKgGpemvjbcTQm7FsbE/GmsLqzR/ONk05VhUyVCs3NWdBJ10VYoNO3tfFAr6jW09lBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T05:39:22.768126Z"},"content_sha256":"9dbe78e2e42f02a34a626fbddf8b97c8c9948a07c56d94746cdcf987ae7d0000","schema_version":"1.0","event_id":"sha256:9dbe78e2e42f02a34a626fbddf8b97c8c9948a07c56d94746cdcf987ae7d0000"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/bundle.json","state_url":"https://pith.science/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/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":"2026-05-27T05:39:22Z","links":{"resolver":"https://pith.science/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4","bundle":"https://pith.science/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/bundle.json","state":"https://pith.science/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MXSZ36Q3R5V43FCS5HBFPUOGI4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MXSZ36Q3R5V43FCS5HBFPUOGI4","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":"ebf386d7d4e2fb534079894272145c9d8a4da7614cec862474d8a7f5c6b3a962","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T05:40:52Z","title_canon_sha256":"fdc0cfc9f785a4663754bbf9c239908277ba75e1c3f43a06d34ce0ec46461fe9"},"schema_version":"1.0","source":{"id":"2605.17879","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17879","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17879v1","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17879","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_12","alias_value":"MXSZ36Q3R5V4","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_16","alias_value":"MXSZ36Q3R5V43FCS","created_at":"2026-05-20T00:05:03Z"},{"alias_kind":"pith_short_8","alias_value":"MXSZ36Q3","created_at":"2026-05-20T00:05:03Z"}],"graph_snapshots":[{"event_id":"sha256:9dbe78e2e42f02a34a626fbddf8b97c8c9948a07c56d94746cdcf987ae7d0000","target":"graph","created_at":"2026-05-20T00:05:03Z","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/2605.17879/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training frontier-scale foundation models involves coordinating tens of thousands of GPUs over multi-month runs, where even minor performance degradations can accumulate into substantial efficiency losses. Existing health-check mechanisms, such as NCCL tests or GPU burn-in, primarily focus on functional correctness and often fail to detect fail-slow behaviors that silently degrade system performance. In this paper, we present Guard, a scalable system for detecting stragglers and ensuring node health in large-scale training clusters. Guard combines lightweight online performance monitoring duri","authors_text":"Abhinandan Patni, Alexander Zhipa, Anthony Ko, Ashvin Nihalani, Binxuan Huang, Cong Cheng, Congzhu Lin, Guanliang Liu, Jack Wittmayer, Josh Wu, Mi Sun, Parthasarathy Govindarajen, Rejith George Joseph, Rory Na, Vijay Rajakumar, Yinghong Liu, Zoe Zeng","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T05:40:52Z","title":"Guard: Scalable Straggler Detection and Node Health Management for Large-Scale Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.17879","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":"sha256:21c489d587b871f037a7939566a57cb32b4c28d82c2620c809696d9796577419","target":"record","created_at":"2026-05-20T00:05:03Z","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":"ebf386d7d4e2fb534079894272145c9d8a4da7614cec862474d8a7f5c6b3a962","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2026-05-18T05:40:52Z","title_canon_sha256":"fdc0cfc9f785a4663754bbf9c239908277ba75e1c3f43a06d34ce0ec46461fe9"},"schema_version":"1.0","source":{"id":"2605.17879","kind":"arxiv","version":1}},"canonical_sha256":"65e59dfa1b8f6bcd9452e9c257d1c6472e64efa35fc500ce7a1a36979da4fffd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"65e59dfa1b8f6bcd9452e9c257d1c6472e64efa35fc500ce7a1a36979da4fffd","first_computed_at":"2026-05-20T00:05:03.295563Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:05:03.295563Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"R/OYpIaFpBlz2VfZ4HWray4MFQbFoNBTKxbJ36ZLiNkfny9vPmbC2+IZ7pFrnNFFbrFOVHsz4jX9eW+jaDQXCw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:05:03.296498Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.17879","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:21c489d587b871f037a7939566a57cb32b4c28d82c2620c809696d9796577419","sha256:9dbe78e2e42f02a34a626fbddf8b97c8c9948a07c56d94746cdcf987ae7d0000"],"state_sha256":"58c28297a6d45a34b1bdce0057ee1a4d5b3d9acfb39483b6b3191fb2932ffeb1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7rPtexfkH47UnZiu/RCrJwG1bCYcGGpQGxnBeYOvR4TNPMuXgMRnxxw/9if8ZLwQPuoXyNbP+QFbl+SbEX4nAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T05:39:22.771567Z","bundle_sha256":"de65f64eedec649b631b7d3e00f0d95ecd19338c7c635576fdae794d147ef708"}}