{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:JALUH5OTSAZV5EVF4EFYBEMQJL","short_pith_number":"pith:JALUH5OT","canonical_record":{"source":{"id":"2312.10793","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-17T18:44:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f182218126f52c0ab7ea871bc919f4414e574ab66c3be1a7df7ddce24ac235c3","abstract_canon_sha256":"996b76e5b6342b9218ec4553c8b740847da5329856502d3caeedb15f94383ff7"},"schema_version":"1.0"},"canonical_sha256":"481743f5d390335e92a5e10b8091904af832b500596a92023a72aa488873b0c8","source":{"kind":"arxiv","id":"2312.10793","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.10793","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"arxiv_version","alias_value":"2312.10793v3","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.10793","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_12","alias_value":"JALUH5OTSAZV","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_16","alias_value":"JALUH5OTSAZV5EVF","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_8","alias_value":"JALUH5OT","created_at":"2026-07-05T07:46:28Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:JALUH5OTSAZV5EVF4EFYBEMQJL","target":"record","payload":{"canonical_record":{"source":{"id":"2312.10793","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-17T18:44:26Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f182218126f52c0ab7ea871bc919f4414e574ab66c3be1a7df7ddce24ac235c3","abstract_canon_sha256":"996b76e5b6342b9218ec4553c8b740847da5329856502d3caeedb15f94383ff7"},"schema_version":"1.0"},"canonical_sha256":"481743f5d390335e92a5e10b8091904af832b500596a92023a72aa488873b0c8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:46:28.913887Z","signature_b64":"0lq6m2oAi2A8G0nurHLgXw0EZt7U+FraSvg2OwQcRILhv0eoEX8TfnbEIxgZ+D4PFb2umZRkw07RZ6XmZMGbBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"481743f5d390335e92a5e10b8091904af832b500596a92023a72aa488873b0c8","last_reissued_at":"2026-07-05T07:46:28.913389Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:46:28.913389Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.10793","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:46:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gQDLH+fSsH7BMqUT1HPFkihnuBQbN13xRdr+oC7HtIzgD9IVgFlYQ03V/ndsIgwRVGsagIbZJ8W7k0d+mQFwAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:54:45.990719Z"},"content_sha256":"4ba9bd0543bfea6528b6940543dc247d38b9aadd55aa508a3e4e009189bbe853","schema_version":"1.0","event_id":"sha256:4ba9bd0543bfea6528b6940543dc247d38b9aadd55aa508a3e4e009189bbe853"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:JALUH5OTSAZV5EVF4EFYBEMQJL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Demystifying Instruction Mixing for Fine-tuning Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chiyu Zhang, Haonan Li, Minghao Wu, Renxi Wang, Timothy Baldwin, Xudong Han, Yuxia Wang","submitted_at":"2023-12-17T18:44:26Z","abstract_excerpt":"Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.10793","kind":"arxiv","version":3},"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/2312.10793/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":"2026-07-05T07:46:28Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"inF9/Slr3nRmn0MlmsE3Ka2rQcF0HxpWP99594ekR104/o7UmbKZHORU5XeuPiJgJMT8mHlbXyqvMyDygC5NBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:54:45.991633Z"},"content_sha256":"7f143602710046401fc98c43bb4d28aaeac9992e38f89ef4d1d07b892a62a7e2","schema_version":"1.0","event_id":"sha256:7f143602710046401fc98c43bb4d28aaeac9992e38f89ef4d1d07b892a62a7e2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/bundle.json","state_url":"https://pith.science/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/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":"2026-08-03T16:54:45Z","links":{"resolver":"https://pith.science/pith/JALUH5OTSAZV5EVF4EFYBEMQJL","bundle":"https://pith.science/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/bundle.json","state":"https://pith.science/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JALUH5OTSAZV5EVF4EFYBEMQJL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:JALUH5OTSAZV5EVF4EFYBEMQJL","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":"996b76e5b6342b9218ec4553c8b740847da5329856502d3caeedb15f94383ff7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-17T18:44:26Z","title_canon_sha256":"f182218126f52c0ab7ea871bc919f4414e574ab66c3be1a7df7ddce24ac235c3"},"schema_version":"1.0","source":{"id":"2312.10793","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.10793","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"arxiv_version","alias_value":"2312.10793v3","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.10793","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_12","alias_value":"JALUH5OTSAZV","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_16","alias_value":"JALUH5OTSAZV5EVF","created_at":"2026-07-05T07:46:28Z"},{"alias_kind":"pith_short_8","alias_value":"JALUH5OT","created_at":"2026-07-05T07:46:28Z"}],"graph_snapshots":[{"event_id":"sha256:7f143602710046401fc98c43bb4d28aaeac9992e38f89ef4d1d07b892a62a7e2","target":"graph","created_at":"2026-07-05T07:46:28Z","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/2312.10793/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights ","authors_text":"Chiyu Zhang, Haonan Li, Minghao Wu, Renxi Wang, Timothy Baldwin, Xudong Han, Yuxia Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-17T18:44:26Z","title":"Demystifying Instruction Mixing for Fine-tuning Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.10793","kind":"arxiv","version":3},"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":"sha256:4ba9bd0543bfea6528b6940543dc247d38b9aadd55aa508a3e4e009189bbe853","target":"record","created_at":"2026-07-05T07:46:28Z","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":"996b76e5b6342b9218ec4553c8b740847da5329856502d3caeedb15f94383ff7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-17T18:44:26Z","title_canon_sha256":"f182218126f52c0ab7ea871bc919f4414e574ab66c3be1a7df7ddce24ac235c3"},"schema_version":"1.0","source":{"id":"2312.10793","kind":"arxiv","version":3}},"canonical_sha256":"481743f5d390335e92a5e10b8091904af832b500596a92023a72aa488873b0c8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"481743f5d390335e92a5e10b8091904af832b500596a92023a72aa488873b0c8","first_computed_at":"2026-07-05T07:46:28.913389Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:46:28.913389Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0lq6m2oAi2A8G0nurHLgXw0EZt7U+FraSvg2OwQcRILhv0eoEX8TfnbEIxgZ+D4PFb2umZRkw07RZ6XmZMGbBg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:46:28.913887Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.10793","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ba9bd0543bfea6528b6940543dc247d38b9aadd55aa508a3e4e009189bbe853","sha256:7f143602710046401fc98c43bb4d28aaeac9992e38f89ef4d1d07b892a62a7e2"],"state_sha256":"d7dbef4ffa91546e40cda532b2b3eb556cc9286d833393a657c539bf77230243"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5nd73JAoiPrFShLfhjz6g9Jd+DAQAhY2KXNcWNUK8hDDVCw2Xs9HHCnmRmrrMNuMX6O/nytHFN7UBaf47FxkAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:54:45.999453Z","bundle_sha256":"4018dc21dd0e044bd8c61c2dddb57af4a234cd96b212f1f459d50a14ad5f820f"}}