{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KBLRZ4FHKHFYXCPCWAFYDM3SGC","short_pith_number":"pith:KBLRZ4FH","canonical_record":{"source":{"id":"2503.21023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-26T22:19:47Z","cross_cats_sorted":[],"title_canon_sha256":"dbc36955f3a7e82610e80defed6bfa1136468598d24b47cd2849c498e84c1f23","abstract_canon_sha256":"1abc799fad4d4d5f8df01371fd7069cd6076481de791202c961afd9ffb2ba69c"},"schema_version":"1.0"},"canonical_sha256":"50571cf0a751cb8b89e2b00b81b37230a74147b66bf87344453b0b8bb90a49aa","source":{"kind":"arxiv","id":"2503.21023","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21023","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21023v1","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21023","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_12","alias_value":"KBLRZ4FHKHFY","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_16","alias_value":"KBLRZ4FHKHFYXCPC","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_8","alias_value":"KBLRZ4FH","created_at":"2026-07-05T10:39:51Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KBLRZ4FHKHFYXCPCWAFYDM3SGC","target":"record","payload":{"canonical_record":{"source":{"id":"2503.21023","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-26T22:19:47Z","cross_cats_sorted":[],"title_canon_sha256":"dbc36955f3a7e82610e80defed6bfa1136468598d24b47cd2849c498e84c1f23","abstract_canon_sha256":"1abc799fad4d4d5f8df01371fd7069cd6076481de791202c961afd9ffb2ba69c"},"schema_version":"1.0"},"canonical_sha256":"50571cf0a751cb8b89e2b00b81b37230a74147b66bf87344453b0b8bb90a49aa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:51.900981Z","signature_b64":"Ga++qSrea+WvcK8GCxvJhkBLpopYZCVctQkFKRHWcUkMgCC2cQxCFqDxErACLqfc58HDLityA1G50Zsbe/V+CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50571cf0a751cb8b89e2b00b81b37230a74147b66bf87344453b0b8bb90a49aa","last_reissued_at":"2026-07-05T10:39:51.900504Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:51.900504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.21023","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":"2026-07-05T10:39:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jmbsnnCk6ZmUL7mrxjks00dmBU6hTfRnYe1FCaEW+uR06PayQ8NH8qsY4rpB99vFUkkg+bmTMahIGpep2wCzCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T06:16:02.705173Z"},"content_sha256":"4e58178c8defb26e9dda12fdba720387ec2f215f33bb3bcae87a289f262179f6","schema_version":"1.0","event_id":"sha256:4e58178c8defb26e9dda12fdba720387ec2f215f33bb3bcae87a289f262179f6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KBLRZ4FHKHFYXCPCWAFYDM3SGC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data Mixture Optimization: A Multi-fidelity Multi-scale Bayesian Framework","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Andrew Wei Tung Siah, Daniel Guetta, Haozhe Chen, Hongseok Namkoong, Thomson Yen, Tianyi Peng","submitted_at":"2025-03-26T22:19:47Z","abstract_excerpt":"Careful curation of data sources can significantly improve the performance of LLM pre-training, but predominant approaches rely heavily on intuition or costly trial-and-error, making them difficult to generalize across different data domains and downstream tasks. Although scaling laws can provide a principled and general approach for data curation, standard deterministic extrapolation from small-scale experiments to larger scales requires strong assumptions on the reliability of such extrapolation, whose brittleness has been highlighted in prior works. In this paper, we introduce a $\\textit{pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21023","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/2503.21023/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":"2026-07-05T10:39:51Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lkwjbWrt3t+yHPvjhAUKSROG/CxfQKy+fygoXfaaqwvMA5wgVhTSUx86ksrcrAWHBtL4gwl53CHqyHO6g5ucBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-11T06:16:02.705833Z"},"content_sha256":"6dbacf8fe71017e7bd604502e61825f664b9f76c3f6e45c2fcc0df606253457b","schema_version":"1.0","event_id":"sha256:6dbacf8fe71017e7bd604502e61825f664b9f76c3f6e45c2fcc0df606253457b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/bundle.json","state_url":"https://pith.science/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/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":"2026-08-11T06:16:02Z","links":{"resolver":"https://pith.science/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC","bundle":"https://pith.science/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/bundle.json","state":"https://pith.science/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KBLRZ4FHKHFYXCPCWAFYDM3SGC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KBLRZ4FHKHFYXCPCWAFYDM3SGC","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":"1abc799fad4d4d5f8df01371fd7069cd6076481de791202c961afd9ffb2ba69c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-26T22:19:47Z","title_canon_sha256":"dbc36955f3a7e82610e80defed6bfa1136468598d24b47cd2849c498e84c1f23"},"schema_version":"1.0","source":{"id":"2503.21023","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.21023","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"arxiv_version","alias_value":"2503.21023v1","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.21023","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_12","alias_value":"KBLRZ4FHKHFY","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_16","alias_value":"KBLRZ4FHKHFYXCPC","created_at":"2026-07-05T10:39:51Z"},{"alias_kind":"pith_short_8","alias_value":"KBLRZ4FH","created_at":"2026-07-05T10:39:51Z"}],"graph_snapshots":[{"event_id":"sha256:6dbacf8fe71017e7bd604502e61825f664b9f76c3f6e45c2fcc0df606253457b","target":"graph","created_at":"2026-07-05T10:39:51Z","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/2503.21023/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Careful curation of data sources can significantly improve the performance of LLM pre-training, but predominant approaches rely heavily on intuition or costly trial-and-error, making them difficult to generalize across different data domains and downstream tasks. Although scaling laws can provide a principled and general approach for data curation, standard deterministic extrapolation from small-scale experiments to larger scales requires strong assumptions on the reliability of such extrapolation, whose brittleness has been highlighted in prior works. In this paper, we introduce a $\\textit{pr","authors_text":"Andrew Wei Tung Siah, Daniel Guetta, Haozhe Chen, Hongseok Namkoong, Thomson Yen, Tianyi Peng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-26T22:19:47Z","title":"Data Mixture Optimization: A Multi-fidelity Multi-scale Bayesian Framework"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.21023","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":"sha256:4e58178c8defb26e9dda12fdba720387ec2f215f33bb3bcae87a289f262179f6","target":"record","created_at":"2026-07-05T10:39:51Z","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":"1abc799fad4d4d5f8df01371fd7069cd6076481de791202c961afd9ffb2ba69c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-26T22:19:47Z","title_canon_sha256":"dbc36955f3a7e82610e80defed6bfa1136468598d24b47cd2849c498e84c1f23"},"schema_version":"1.0","source":{"id":"2503.21023","kind":"arxiv","version":1}},"canonical_sha256":"50571cf0a751cb8b89e2b00b81b37230a74147b66bf87344453b0b8bb90a49aa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"50571cf0a751cb8b89e2b00b81b37230a74147b66bf87344453b0b8bb90a49aa","first_computed_at":"2026-07-05T10:39:51.900504Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:39:51.900504Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ga++qSrea+WvcK8GCxvJhkBLpopYZCVctQkFKRHWcUkMgCC2cQxCFqDxErACLqfc58HDLityA1G50Zsbe/V+CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:39:51.900981Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.21023","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4e58178c8defb26e9dda12fdba720387ec2f215f33bb3bcae87a289f262179f6","sha256:6dbacf8fe71017e7bd604502e61825f664b9f76c3f6e45c2fcc0df606253457b"],"state_sha256":"9340612453f7de78e9c35ba865dcd1a0c74f749dc4d0bfe023bc7314d454a8d6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E1HG4jBtCQaw/V4KwTa6u3DmnGiZ6QSa5ipc6F6wdgljek0MssBtSc5XUUkx6CJqsfB+EFy4JWS6Sw1oj/vADg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-11T06:16:02.711699Z","bundle_sha256":"b1a9037a7e62e8884ca7b7b4eb20989de3547a5109de425f243ed0ad7c5e2cad"}}