{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LQRU3AGDMEZZB7JKQHVPLU3N34","short_pith_number":"pith:LQRU3AGD","canonical_record":{"source":{"id":"2505.01043","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T06:33:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b7efb361c58702f2f0d63efd26e0e86e1709d5abafc1f5c326f4a7b0d42c8936","abstract_canon_sha256":"d0815a4c77fee47e286c33ffb31ed7dcf63a40871089ee1c17e74554ad490dfb"},"schema_version":"1.0"},"canonical_sha256":"5c234d80c3613390fd2a81eaf5d36ddf2e410f547dc8f45e94c899eb8e3c6dee","source":{"kind":"arxiv","id":"2505.01043","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01043","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01043v1","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01043","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_12","alias_value":"LQRU3AGDMEZZ","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_16","alias_value":"LQRU3AGDMEZZB7JK","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_8","alias_value":"LQRU3AGD","created_at":"2026-07-05T10:57:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LQRU3AGDMEZZB7JKQHVPLU3N34","target":"record","payload":{"canonical_record":{"source":{"id":"2505.01043","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T06:33:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b7efb361c58702f2f0d63efd26e0e86e1709d5abafc1f5c326f4a7b0d42c8936","abstract_canon_sha256":"d0815a4c77fee47e286c33ffb31ed7dcf63a40871089ee1c17e74554ad490dfb"},"schema_version":"1.0"},"canonical_sha256":"5c234d80c3613390fd2a81eaf5d36ddf2e410f547dc8f45e94c899eb8e3c6dee","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:57:44.042046Z","signature_b64":"n4pDiX5CmXwaZ1UpgOLI4KLW4mLYkuz1rjEwX4HqZS3hUHuKL9UrE2z2QN04J/48IwYM23HEM1zJ2O4w6D6SDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5c234d80c3613390fd2a81eaf5d36ddf2e410f547dc8f45e94c899eb8e3c6dee","last_reissued_at":"2026-07-05T10:57:44.041299Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:57:44.041299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.01043","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":"2026-07-05T10:57:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GKFL6iZG0G8X1dLw29x5GUt0QuCjYVRhPVp3p/LAxJV0apcGOJf9yVSf/YXqJsc0XNyz2vtbrPoa4uAXicXvCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T01:49:10.982149Z"},"content_sha256":"905bf50244d2d5a7f5b5c38af1e4e963aa5a87033aacb4bd9a9799ca828cfa3a","schema_version":"1.0","event_id":"sha256:905bf50244d2d5a7f5b5c38af1e4e963aa5a87033aacb4bd9a9799ca828cfa3a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LQRU3AGDMEZZB7JKQHVPLU3N34","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Dacheng Tao, Dianhai Yu, Guoxia Wang, Han Hu, Jianyuan Guo, Li Shen, Yonggang Wen, Yong Luo, Zhiwei Hao","submitted_at":"2025-05-02T06:33:25Z","abstract_excerpt":"Large language models (LLMs) have achieved impressive performance across various domains. However, the substantial hardware resources required for their training present a significant barrier to efficiency and scalability. To mitigate this challenge, low-precision training techniques have been widely adopted, leading to notable advancements in training efficiency. Despite these gains, low-precision training involves several components$\\unicode{x2013}$such as weights, activations, and gradients$\\unicode{x2013}$each of which can be represented in different numerical formats. The resulting divers"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01043","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/2505.01043/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":"2026-07-05T10:57:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wxGuR8sBs2X7Q7+n02WG02V7Qlp0BHxsjtGKr07zgCyp6N8+HDElSOif+ynjrcH0bTpURNPXNskeArbOuxPLBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-20T01:49:10.982732Z"},"content_sha256":"abdb3aef8684928a9c62e4c26b004d04fa195e19677021ba2e81b9013718de2b","schema_version":"1.0","event_id":"sha256:abdb3aef8684928a9c62e4c26b004d04fa195e19677021ba2e81b9013718de2b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/bundle.json","state_url":"https://pith.science/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/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":"2026-08-20T01:49:10Z","links":{"resolver":"https://pith.science/pith/LQRU3AGDMEZZB7JKQHVPLU3N34","bundle":"https://pith.science/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/bundle.json","state":"https://pith.science/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LQRU3AGDMEZZB7JKQHVPLU3N34/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LQRU3AGDMEZZB7JKQHVPLU3N34","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":"d0815a4c77fee47e286c33ffb31ed7dcf63a40871089ee1c17e74554ad490dfb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T06:33:25Z","title_canon_sha256":"b7efb361c58702f2f0d63efd26e0e86e1709d5abafc1f5c326f4a7b0d42c8936"},"schema_version":"1.0","source":{"id":"2505.01043","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.01043","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"arxiv_version","alias_value":"2505.01043v1","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.01043","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_12","alias_value":"LQRU3AGDMEZZ","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_16","alias_value":"LQRU3AGDMEZZB7JK","created_at":"2026-07-05T10:57:44Z"},{"alias_kind":"pith_short_8","alias_value":"LQRU3AGD","created_at":"2026-07-05T10:57:44Z"}],"graph_snapshots":[{"event_id":"sha256:abdb3aef8684928a9c62e4c26b004d04fa195e19677021ba2e81b9013718de2b","target":"graph","created_at":"2026-07-05T10:57:44Z","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/2505.01043/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) have achieved impressive performance across various domains. However, the substantial hardware resources required for their training present a significant barrier to efficiency and scalability. To mitigate this challenge, low-precision training techniques have been widely adopted, leading to notable advancements in training efficiency. Despite these gains, low-precision training involves several components$\\unicode{x2013}$such as weights, activations, and gradients$\\unicode{x2013}$each of which can be represented in different numerical formats. The resulting divers","authors_text":"Dacheng Tao, Dianhai Yu, Guoxia Wang, Han Hu, Jianyuan Guo, Li Shen, Yonggang Wen, Yong Luo, Zhiwei Hao","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T06:33:25Z","title":"Low-Precision Training of Large Language Models: Methods, Challenges, and Opportunities"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.01043","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":"sha256:905bf50244d2d5a7f5b5c38af1e4e963aa5a87033aacb4bd9a9799ca828cfa3a","target":"record","created_at":"2026-07-05T10:57:44Z","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":"d0815a4c77fee47e286c33ffb31ed7dcf63a40871089ee1c17e74554ad490dfb","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-02T06:33:25Z","title_canon_sha256":"b7efb361c58702f2f0d63efd26e0e86e1709d5abafc1f5c326f4a7b0d42c8936"},"schema_version":"1.0","source":{"id":"2505.01043","kind":"arxiv","version":1}},"canonical_sha256":"5c234d80c3613390fd2a81eaf5d36ddf2e410f547dc8f45e94c899eb8e3c6dee","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5c234d80c3613390fd2a81eaf5d36ddf2e410f547dc8f45e94c899eb8e3c6dee","first_computed_at":"2026-07-05T10:57:44.041299Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:57:44.041299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n4pDiX5CmXwaZ1UpgOLI4KLW4mLYkuz1rjEwX4HqZS3hUHuKL9UrE2z2QN04J/48IwYM23HEM1zJ2O4w6D6SDg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:57:44.042046Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.01043","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:905bf50244d2d5a7f5b5c38af1e4e963aa5a87033aacb4bd9a9799ca828cfa3a","sha256:abdb3aef8684928a9c62e4c26b004d04fa195e19677021ba2e81b9013718de2b"],"state_sha256":"d1c7996e63f23814646c5f76521f6f833b97f38b9a6e526052b453c00017cbdf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AFNA2Hi1KYWDg6gyb2d45sx9KQJqDvtOcl/akFFI6ZjcxEtjXBWfuwfNE0f1ozBWst9C0B8PdHclb8Zp56TOCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-20T01:49:10.987401Z","bundle_sha256":"d01a8350a246e1045758ca9df20417daebcf0f956e02c29f37b0ed76083a02b8"}}