{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AVOTLRISYCA6OV7S7ZP46F4AW4","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":"8b0d3660f2f76d384b440a0b834b07e32776b87ed20f66481acc3423e86061ab","cross_cats_sorted":["cs.AI","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:22:14Z","title_canon_sha256":"7f6f2c43c45fff76889f79c087f44515e0690c4c90c1a2b5813c1509cf8be43a"},"schema_version":"1.0","source":{"id":"2411.05059","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.05059","created_at":"2026-07-05T09:34:00Z"},{"alias_kind":"arxiv_version","alias_value":"2411.05059v2","created_at":"2026-07-05T09:34:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.05059","created_at":"2026-07-05T09:34:00Z"},{"alias_kind":"pith_short_12","alias_value":"AVOTLRISYCA6","created_at":"2026-07-05T09:34:00Z"},{"alias_kind":"pith_short_16","alias_value":"AVOTLRISYCA6OV7S","created_at":"2026-07-05T09:34:00Z"},{"alias_kind":"pith_short_8","alias_value":"AVOTLRIS","created_at":"2026-07-05T09:34:00Z"}],"graph_snapshots":[{"event_id":"sha256:4ab8640f0c82e300bd6459f0fb15ba609a6294df40b36391947191eeeb6106f2","target":"graph","created_at":"2026-07-05T09:34:00Z","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/2411.05059/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"There is great interest in fine-tuning frontier large language models (LLMs) to inject new information and update existing knowledge. While commercial LLM fine-tuning APIs from providers such as OpenAI and Google promise flexible adaptation for various applications, the efficacy of fine-tuning remains unclear. In this study, we introduce FineTuneBench, an evaluation framework and dataset for understanding how well commercial fine-tuning APIs can successfully learn new and updated knowledge. We analyze five frontier LLMs with commercially available fine-tuning APIs, including GPT-4o and Gemini ","authors_text":"Eric Wu, James Zou, Kevin Wu","cross_cats":["cs.AI","cs.IR"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:22:14Z","title":"FineTuneBench: How well do commercial fine-tuning APIs infuse knowledge into LLMs?"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.05059","kind":"arxiv","version":2},"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":"sha256:3f22b7088d3ab02159efa8de7811987bb70d638de14ed1c14ae936c4e43ef376","target":"record","created_at":"2026-07-05T09:34:00Z","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":"8b0d3660f2f76d384b440a0b834b07e32776b87ed20f66481acc3423e86061ab","cross_cats_sorted":["cs.AI","cs.IR"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-11-07T18:22:14Z","title_canon_sha256":"7f6f2c43c45fff76889f79c087f44515e0690c4c90c1a2b5813c1509cf8be43a"},"schema_version":"1.0","source":{"id":"2411.05059","kind":"arxiv","version":2}},"canonical_sha256":"055d35c512c081e757f2fe5fcf1780b72959e1b6cbd4ffb8b12d5ea9d2ce4590","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"055d35c512c081e757f2fe5fcf1780b72959e1b6cbd4ffb8b12d5ea9d2ce4590","first_computed_at":"2026-07-05T09:34:00.511473Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:34:00.511473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2a32d1L7Xu7A2/WGmujFaos9hWZwwu/gFt/3rNWuw6kcRJad9tv9Tutk3W6DtarG4f6Se2D3MQZKzn/0I5hSBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:34:00.511917Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.05059","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3f22b7088d3ab02159efa8de7811987bb70d638de14ed1c14ae936c4e43ef376","sha256:4ab8640f0c82e300bd6459f0fb15ba609a6294df40b36391947191eeeb6106f2"],"state_sha256":"97d5ed64d9e94cad32d5ba8e835f603028611a428e278292b1832e4935555939"}