{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:BLVJMANIWTRRP5QPGZVGA76EL3","short_pith_number":"pith:BLVJMANI","canonical_record":{"source":{"id":"2508.11953","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:28:39Z","cross_cats_sorted":[],"title_canon_sha256":"ff1f6c5087657f308faf9ede38a518cf0ec8b664054264817d03e10034e9de98","abstract_canon_sha256":"6e01f89e10e7a28bfe5bc7e6467239daa377f477b666f22cce14e6fd85d3a84a"},"schema_version":"1.0"},"canonical_sha256":"0aea9601a8b4e317f60f366a607fc45ef06baca88dcf3b74121829715967d99f","source":{"kind":"arxiv","id":"2508.11953","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11953","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11953v1","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11953","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_12","alias_value":"BLVJMANIWTRR","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_16","alias_value":"BLVJMANIWTRRP5QP","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_8","alias_value":"BLVJMANI","created_at":"2026-07-05T11:54:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:BLVJMANIWTRRP5QPGZVGA76EL3","target":"record","payload":{"canonical_record":{"source":{"id":"2508.11953","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:28:39Z","cross_cats_sorted":[],"title_canon_sha256":"ff1f6c5087657f308faf9ede38a518cf0ec8b664054264817d03e10034e9de98","abstract_canon_sha256":"6e01f89e10e7a28bfe5bc7e6467239daa377f477b666f22cce14e6fd85d3a84a"},"schema_version":"1.0"},"canonical_sha256":"0aea9601a8b4e317f60f366a607fc45ef06baca88dcf3b74121829715967d99f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:57.745198Z","signature_b64":"jlGNHTYeeXwOokOQvfp4Rffm+kE/bvD5quWqJw8JIJrAsaKgoJofbyaLZXUBJTr+my44FecU5PCSBbGlvWbuDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0aea9601a8b4e317f60f366a607fc45ef06baca88dcf3b74121829715967d99f","last_reissued_at":"2026-07-05T11:54:57.744778Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:57.744778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2508.11953","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":"2026-07-05T11:54:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AIeczWllOeidMdAKh3nSTmEPHgIeSCIwthAgrKy+j5q/pw99N3qN41Az4MIKkCM7rjIqdTLoC2iHJ4z1udx7Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:15:44.083477Z"},"content_sha256":"838aa9a1e027f2cbdb515606e07b3a2ae98b629cc1217f67cd3caf660b54b85a","schema_version":"1.0","event_id":"sha256:838aa9a1e027f2cbdb515606e07b3a2ae98b629cc1217f67cd3caf660b54b85a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:BLVJMANIWTRRP5QPGZVGA76EL3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Data Mixing Optimization for Supervised Fine-Tuning of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Eric Xing, Yuan Li, Zhengzhong Liu","submitted_at":"2025-08-16T07:28:39Z","abstract_excerpt":"Optimizing data mixtures for supervised fine-tuning (SFT) of large language models (LLMs) is critical for developing general-purpose models, yet this area remains underexplored. In this paper, we frame data mixing as an optimization problem and introduce a novel method designed to minimize validation loss. Our approach parametrizes the loss by modeling effective data transferred and leveraging scaling laws for fine-tuning. By experimenting with various small-scale data mixtures, we fit these parameters and derive the optimal weights. We provide both mathematical proofs and empirical results de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11953","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/2508.11953/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":"2026-07-05T11:54:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VxjDgpvb3m1SINYTcVpX7n7Ul/ewF+rfQM0yLSqO2pDfP+YrVaPUJmNx0r0hSyF0pTWEm0/yg+ZANVCrZHJDBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T14:15:44.084002Z"},"content_sha256":"e75a8da1e95f9681e4d07222842af951004cfce0e6e575a0238bf12a7d70ed79","schema_version":"1.0","event_id":"sha256:e75a8da1e95f9681e4d07222842af951004cfce0e6e575a0238bf12a7d70ed79"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BLVJMANIWTRRP5QPGZVGA76EL3/bundle.json","state_url":"https://pith.science/pith/BLVJMANIWTRRP5QPGZVGA76EL3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BLVJMANIWTRRP5QPGZVGA76EL3/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":"2026-08-19T14:15:44Z","links":{"resolver":"https://pith.science/pith/BLVJMANIWTRRP5QPGZVGA76EL3","bundle":"https://pith.science/pith/BLVJMANIWTRRP5QPGZVGA76EL3/bundle.json","state":"https://pith.science/pith/BLVJMANIWTRRP5QPGZVGA76EL3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BLVJMANIWTRRP5QPGZVGA76EL3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:BLVJMANIWTRRP5QPGZVGA76EL3","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":"6e01f89e10e7a28bfe5bc7e6467239daa377f477b666f22cce14e6fd85d3a84a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:28:39Z","title_canon_sha256":"ff1f6c5087657f308faf9ede38a518cf0ec8b664054264817d03e10034e9de98"},"schema_version":"1.0","source":{"id":"2508.11953","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.11953","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"arxiv_version","alias_value":"2508.11953v1","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11953","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_12","alias_value":"BLVJMANIWTRR","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_16","alias_value":"BLVJMANIWTRRP5QP","created_at":"2026-07-05T11:54:57Z"},{"alias_kind":"pith_short_8","alias_value":"BLVJMANI","created_at":"2026-07-05T11:54:57Z"}],"graph_snapshots":[{"event_id":"sha256:e75a8da1e95f9681e4d07222842af951004cfce0e6e575a0238bf12a7d70ed79","target":"graph","created_at":"2026-07-05T11:54:57Z","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/2508.11953/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Optimizing data mixtures for supervised fine-tuning (SFT) of large language models (LLMs) is critical for developing general-purpose models, yet this area remains underexplored. In this paper, we frame data mixing as an optimization problem and introduce a novel method designed to minimize validation loss. Our approach parametrizes the loss by modeling effective data transferred and leveraging scaling laws for fine-tuning. By experimenting with various small-scale data mixtures, we fit these parameters and derive the optimal weights. We provide both mathematical proofs and empirical results de","authors_text":"Eric Xing, Yuan Li, Zhengzhong Liu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:28:39Z","title":"Data Mixing Optimization for Supervised Fine-Tuning of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11953","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":"sha256:838aa9a1e027f2cbdb515606e07b3a2ae98b629cc1217f67cd3caf660b54b85a","target":"record","created_at":"2026-07-05T11:54:57Z","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":"6e01f89e10e7a28bfe5bc7e6467239daa377f477b666f22cce14e6fd85d3a84a","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-08-16T07:28:39Z","title_canon_sha256":"ff1f6c5087657f308faf9ede38a518cf0ec8b664054264817d03e10034e9de98"},"schema_version":"1.0","source":{"id":"2508.11953","kind":"arxiv","version":1}},"canonical_sha256":"0aea9601a8b4e317f60f366a607fc45ef06baca88dcf3b74121829715967d99f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0aea9601a8b4e317f60f366a607fc45ef06baca88dcf3b74121829715967d99f","first_computed_at":"2026-07-05T11:54:57.744778Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:54:57.744778Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jlGNHTYeeXwOokOQvfp4Rffm+kE/bvD5quWqJw8JIJrAsaKgoJofbyaLZXUBJTr+my44FecU5PCSBbGlvWbuDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:54:57.745198Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.11953","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:838aa9a1e027f2cbdb515606e07b3a2ae98b629cc1217f67cd3caf660b54b85a","sha256:e75a8da1e95f9681e4d07222842af951004cfce0e6e575a0238bf12a7d70ed79"],"state_sha256":"f3b16d0ca62a55b92406d855f2e9a651ec38a9f2c197ddf68a05344eb453f0ef"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uqsHrRiCUnBFHm4Q/M1ohEqKaNR+pxDOf7otuHMuj/413MRORcqdq17dmtGGxA+nlXtuHAOzQsge6qv+gai2BA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T14:15:44.089865Z","bundle_sha256":"dfa6dc5cd642eaf7ac960ca60a474430a6cd766d178383fe98206368e64fed6e"}}