{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:FOMZARMHPAWPNK5CFBRZXUUMGT","short_pith_number":"pith:FOMZARMH","canonical_record":{"source":{"id":"2505.11170","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T12:14:12Z","cross_cats_sorted":[],"title_canon_sha256":"50fd4168ed7ee261aac88ea3b044d417eaa52caeef111343e1a00b33aabcbd12","abstract_canon_sha256":"26236c73fae7f82711adabdfe39b1087d1bf84195c0a8ad6241f55fac07a43d8"},"schema_version":"1.0"},"canonical_sha256":"2b99904587782cf6aba228639bd28c34cb802852e01bbf4d2873063b336ae7d0","source":{"kind":"arxiv","id":"2505.11170","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11170","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11170v1","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11170","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"FOMZARMHPAWP","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"FOMZARMHPAWPNK5C","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"FOMZARMH","created_at":"2026-07-05T11:04:12Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:FOMZARMHPAWPNK5CFBRZXUUMGT","target":"record","payload":{"canonical_record":{"source":{"id":"2505.11170","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T12:14:12Z","cross_cats_sorted":[],"title_canon_sha256":"50fd4168ed7ee261aac88ea3b044d417eaa52caeef111343e1a00b33aabcbd12","abstract_canon_sha256":"26236c73fae7f82711adabdfe39b1087d1bf84195c0a8ad6241f55fac07a43d8"},"schema_version":"1.0"},"canonical_sha256":"2b99904587782cf6aba228639bd28c34cb802852e01bbf4d2873063b336ae7d0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:12.099056Z","signature_b64":"VionWIhc4h9j1MWbSXyYrRX/ioatL2YR95N4SMZuOUVwRf1jhnNgVPNu49kQbksR2pmPyr/LkjhssZTmNWFPCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b99904587782cf6aba228639bd28c34cb802852e01bbf4d2873063b336ae7d0","last_reissued_at":"2026-07-05T11:04:12.098611Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:12.098611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.11170","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":"2026-07-05T11:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nB/4yYtrrbKu5L6yVsEzktCGfgqshzOuHwphedTMk4flyHYpHzxITv7BhjQlKQzkTZiqghqckJvvrfAcylpADw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T05:10:33.835947Z"},"content_sha256":"a6e6da9872985f86dcd0b0fabec61180d6d46f5ba17c5f68be29f0b6071fdd03","schema_version":"1.0","event_id":"sha256:a6e6da9872985f86dcd0b0fabec61180d6d46f5ba17c5f68be29f0b6071fdd03"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:FOMZARMHPAWPNK5CFBRZXUUMGT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Myeonghwan Ahn, Sungjoo Yoo","submitted_at":"2025-05-16T12:14:12Z","abstract_excerpt":"Ever-growing scale of large language models (LLMs) is pushing for improved efficiency, favoring fully quantized training (FQT) over BF16. While FQT accelerates training, it faces consistency challenges and requires searching over an exponential number of cases, each needing over 200B tokens to ensure stability.\n  Pseudo-quantization training (PQT) addresses the issues of FQT, although it is not well-studied. We explore the practical implications of PQT in detail and propose a noise distribution $R$ that is floating-point (FP)-friendly, with ideal properties including stochastic precision annea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11170","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/2505.11170/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":"2026-07-05T11:04:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SFwLoW717Lu+rmEk38k1HPb+G127TtmDSHWFtOQIXIbcV4r+EOkfNB+vLJtCGRSFWa13xsbhW+rzujgLK6k2CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T05:10:33.857002Z"},"content_sha256":"48a367addf51eaa86a887de5047447468657336e4317c208eec316b33263455f","schema_version":"1.0","event_id":"sha256:48a367addf51eaa86a887de5047447468657336e4317c208eec316b33263455f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/bundle.json","state_url":"https://pith.science/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/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":"2026-08-16T05:10:33Z","links":{"resolver":"https://pith.science/pith/FOMZARMHPAWPNK5CFBRZXUUMGT","bundle":"https://pith.science/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/bundle.json","state":"https://pith.science/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FOMZARMHPAWPNK5CFBRZXUUMGT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:FOMZARMHPAWPNK5CFBRZXUUMGT","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":"26236c73fae7f82711adabdfe39b1087d1bf84195c0a8ad6241f55fac07a43d8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T12:14:12Z","title_canon_sha256":"50fd4168ed7ee261aac88ea3b044d417eaa52caeef111343e1a00b33aabcbd12"},"schema_version":"1.0","source":{"id":"2505.11170","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.11170","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"arxiv_version","alias_value":"2505.11170v1","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11170","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_12","alias_value":"FOMZARMHPAWP","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_16","alias_value":"FOMZARMHPAWPNK5C","created_at":"2026-07-05T11:04:12Z"},{"alias_kind":"pith_short_8","alias_value":"FOMZARMH","created_at":"2026-07-05T11:04:12Z"}],"graph_snapshots":[{"event_id":"sha256:48a367addf51eaa86a887de5047447468657336e4317c208eec316b33263455f","target":"graph","created_at":"2026-07-05T11:04:12Z","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/2505.11170/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Ever-growing scale of large language models (LLMs) is pushing for improved efficiency, favoring fully quantized training (FQT) over BF16. While FQT accelerates training, it faces consistency challenges and requires searching over an exponential number of cases, each needing over 200B tokens to ensure stability.\n  Pseudo-quantization training (PQT) addresses the issues of FQT, although it is not well-studied. We explore the practical implications of PQT in detail and propose a noise distribution $R$ that is floating-point (FP)-friendly, with ideal properties including stochastic precision annea","authors_text":"Myeonghwan Ahn, Sungjoo Yoo","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T12:14:12Z","title":"Gaussian Weight Sampling for Scalable, Efficient and Stable Pseudo-Quantization Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11170","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:a6e6da9872985f86dcd0b0fabec61180d6d46f5ba17c5f68be29f0b6071fdd03","target":"record","created_at":"2026-07-05T11:04:12Z","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":"26236c73fae7f82711adabdfe39b1087d1bf84195c0a8ad6241f55fac07a43d8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T12:14:12Z","title_canon_sha256":"50fd4168ed7ee261aac88ea3b044d417eaa52caeef111343e1a00b33aabcbd12"},"schema_version":"1.0","source":{"id":"2505.11170","kind":"arxiv","version":1}},"canonical_sha256":"2b99904587782cf6aba228639bd28c34cb802852e01bbf4d2873063b336ae7d0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2b99904587782cf6aba228639bd28c34cb802852e01bbf4d2873063b336ae7d0","first_computed_at":"2026-07-05T11:04:12.098611Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:04:12.098611Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VionWIhc4h9j1MWbSXyYrRX/ioatL2YR95N4SMZuOUVwRf1jhnNgVPNu49kQbksR2pmPyr/LkjhssZTmNWFPCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:04:12.099056Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.11170","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a6e6da9872985f86dcd0b0fabec61180d6d46f5ba17c5f68be29f0b6071fdd03","sha256:48a367addf51eaa86a887de5047447468657336e4317c208eec316b33263455f"],"state_sha256":"046006593ad308027de44e438b9c16d39712458f28cca71486233ef1b26fded8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+pxLKAc7bmd2VQi4BsPq1u8X+FxIs5Aa31kLd5AF4RIiErN+UyEYJ+aVDV85PxzYL+FfJLx+AZjPi+sF3fDLDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T05:10:33.933939Z","bundle_sha256":"f357ae1c061231e5127a614bf902bcb4bbab6fb8d182910522a5bd76f728daba"}}