{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:RJFZVYLJ3WT5LJ3YS5EPVOKUBA","short_pith_number":"pith:RJFZVYLJ","canonical_record":{"source":{"id":"2403.12776","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T14:44:45Z","cross_cats_sorted":[],"title_canon_sha256":"a9fa2cc6ac743191627f511d3a76f37d6ee969092cc6343adbc9d58b11bb9f37","abstract_canon_sha256":"02430b20e06d0117ba3accafa3017e2aea5807fa8192ba460d89e190a4acc6b2"},"schema_version":"1.0"},"canonical_sha256":"8a4b9ae169dda7d5a7789748fab954080fb01dea5a4e5ab8492a5520ab2a1381","source":{"kind":"arxiv","id":"2403.12776","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12776","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12776v1","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12776","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_12","alias_value":"RJFZVYLJ3WT5","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_16","alias_value":"RJFZVYLJ3WT5LJ3Y","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_8","alias_value":"RJFZVYLJ","created_at":"2026-07-05T07:58:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:RJFZVYLJ3WT5LJ3YS5EPVOKUBA","target":"record","payload":{"canonical_record":{"source":{"id":"2403.12776","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T14:44:45Z","cross_cats_sorted":[],"title_canon_sha256":"a9fa2cc6ac743191627f511d3a76f37d6ee969092cc6343adbc9d58b11bb9f37","abstract_canon_sha256":"02430b20e06d0117ba3accafa3017e2aea5807fa8192ba460d89e190a4acc6b2"},"schema_version":"1.0"},"canonical_sha256":"8a4b9ae169dda7d5a7789748fab954080fb01dea5a4e5ab8492a5520ab2a1381","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:58:06.926437Z","signature_b64":"fwyobkVCRVt6NrBtYIr9/Vq4Gc0W6gRe9QMAigG5jaV+dIi+W1Zrbhfjdem3vuauX0gVJS515KEsh373PNdADw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8a4b9ae169dda7d5a7789748fab954080fb01dea5a4e5ab8492a5520ab2a1381","last_reissued_at":"2026-07-05T07:58:06.925836Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:58:06.925836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.12776","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":"2026-07-05T07:58:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q/sHiBhHhMWutQPSFvLKWlgwwCWCU/SXzCISCtX6jmyqVuyB8PwUZm0QaaiSXpuPiKMlhhgmRrfNuGaJvQyNDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T17:25:51.644256Z"},"content_sha256":"6d8161796610fe4d69bf65247fe591b135fd952ccaa3b5cb18da127769465ff4","schema_version":"1.0","event_id":"sha256:6d8161796610fe4d69bf65247fe591b135fd952ccaa3b5cb18da127769465ff4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:RJFZVYLJ3WT5LJ3YS5EPVOKUBA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Automated Data Curation for Robust Language Model Fine-Tuning","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiuhai Chen, Jonas Mueller","submitted_at":"2024-03-19T14:44:45Z","abstract_excerpt":"Large Language Models have become the de facto approach to sequence-to-sequence text generation tasks, but for specialized tasks/domains, a pretrained LLM lacks specific capabilities to produce accurate or well-formatted responses. Supervised fine-tuning specializes a LLM by training it on dataset of example prompts with target responses, but real-world data tends to be noisy. While many fine-tuning algorithms exist, here we consider a \\emph{data-centric AI} perspective on LLM fine-tuning, studying how to \\emph{systematically} curate the training dataset to improve the LLM produced via \\emph{a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12776","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/2403.12776/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":"2026-07-05T07:58:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+qNYtxsh9FgWMF6gMolgl5kcCZqWWQRCPkn00ph5aoWlCwi/JTFVr2zrkFBWOHThSnPaQXg+BZppQcthhQ0kCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T17:25:51.645250Z"},"content_sha256":"82976fb08096526e42c742114924b5154e66e52359064dc36398e5d3b50c01fb","schema_version":"1.0","event_id":"sha256:82976fb08096526e42c742114924b5154e66e52359064dc36398e5d3b50c01fb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/bundle.json","state_url":"https://pith.science/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/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":"2026-08-12T17:25:51Z","links":{"resolver":"https://pith.science/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA","bundle":"https://pith.science/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/bundle.json","state":"https://pith.science/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RJFZVYLJ3WT5LJ3YS5EPVOKUBA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:RJFZVYLJ3WT5LJ3YS5EPVOKUBA","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":"02430b20e06d0117ba3accafa3017e2aea5807fa8192ba460d89e190a4acc6b2","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T14:44:45Z","title_canon_sha256":"a9fa2cc6ac743191627f511d3a76f37d6ee969092cc6343adbc9d58b11bb9f37"},"schema_version":"1.0","source":{"id":"2403.12776","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12776","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12776v1","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12776","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_12","alias_value":"RJFZVYLJ3WT5","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_16","alias_value":"RJFZVYLJ3WT5LJ3Y","created_at":"2026-07-05T07:58:06Z"},{"alias_kind":"pith_short_8","alias_value":"RJFZVYLJ","created_at":"2026-07-05T07:58:06Z"}],"graph_snapshots":[{"event_id":"sha256:82976fb08096526e42c742114924b5154e66e52359064dc36398e5d3b50c01fb","target":"graph","created_at":"2026-07-05T07:58:06Z","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/2403.12776/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Models have become the de facto approach to sequence-to-sequence text generation tasks, but for specialized tasks/domains, a pretrained LLM lacks specific capabilities to produce accurate or well-formatted responses. Supervised fine-tuning specializes a LLM by training it on dataset of example prompts with target responses, but real-world data tends to be noisy. While many fine-tuning algorithms exist, here we consider a \\emph{data-centric AI} perspective on LLM fine-tuning, studying how to \\emph{systematically} curate the training dataset to improve the LLM produced via \\emph{a","authors_text":"Jiuhai Chen, Jonas Mueller","cross_cats":[],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T14:44:45Z","title":"Automated Data Curation for Robust Language Model Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12776","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":"sha256:6d8161796610fe4d69bf65247fe591b135fd952ccaa3b5cb18da127769465ff4","target":"record","created_at":"2026-07-05T07:58:06Z","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":"02430b20e06d0117ba3accafa3017e2aea5807fa8192ba460d89e190a4acc6b2","cross_cats_sorted":[],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2024-03-19T14:44:45Z","title_canon_sha256":"a9fa2cc6ac743191627f511d3a76f37d6ee969092cc6343adbc9d58b11bb9f37"},"schema_version":"1.0","source":{"id":"2403.12776","kind":"arxiv","version":1}},"canonical_sha256":"8a4b9ae169dda7d5a7789748fab954080fb01dea5a4e5ab8492a5520ab2a1381","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"8a4b9ae169dda7d5a7789748fab954080fb01dea5a4e5ab8492a5520ab2a1381","first_computed_at":"2026-07-05T07:58:06.925836Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:58:06.925836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fwyobkVCRVt6NrBtYIr9/Vq4Gc0W6gRe9QMAigG5jaV+dIi+W1Zrbhfjdem3vuauX0gVJS515KEsh373PNdADw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:58:06.926437Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.12776","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d8161796610fe4d69bf65247fe591b135fd952ccaa3b5cb18da127769465ff4","sha256:82976fb08096526e42c742114924b5154e66e52359064dc36398e5d3b50c01fb"],"state_sha256":"25ab9b75ebdc06a60bd297dbaecb9dbe760e99486c83bb7260b485ae4b8a7af3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u9rmcsAq6zFiwhboqxh0d/qj9SxOOPADkQhZIiP59WoppuUYm4I2H8fR9/qQbpmd1acZSc5qL2Y2R8X2bZFkDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T17:25:51.651538Z","bundle_sha256":"347886b0f1ec6f2061172651dd4d38a78ff16603fb6ad0993850489874e39472"}}