{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:H2AMZNXSRQJZCCVTEJAG7R7DSP","short_pith_number":"pith:H2AMZNXS","canonical_record":{"source":{"id":"2410.03743","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-01T08:44:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"796373efc50b7a50220c7f5fb6b68afe2ba58888c28ef2839f4f35f77506155b","abstract_canon_sha256":"dd72ac23c33bf06abf94c2dfa1233308d6b587a2a8e8014c7affb26d4362313c"},"schema_version":"1.0"},"canonical_sha256":"3e80ccb6f28c13910ab322406fc7e393ec0dd06d4573c958aadd2b6f3fa62fa0","source":{"kind":"arxiv","id":"2410.03743","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03743","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03743v1","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03743","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_12","alias_value":"H2AMZNXSRQJZ","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_16","alias_value":"H2AMZNXSRQJZCCVT","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_8","alias_value":"H2AMZNXS","created_at":"2026-07-05T09:16:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:H2AMZNXSRQJZCCVTEJAG7R7DSP","target":"record","payload":{"canonical_record":{"source":{"id":"2410.03743","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-01T08:44:31Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"796373efc50b7a50220c7f5fb6b68afe2ba58888c28ef2839f4f35f77506155b","abstract_canon_sha256":"dd72ac23c33bf06abf94c2dfa1233308d6b587a2a8e8014c7affb26d4362313c"},"schema_version":"1.0"},"canonical_sha256":"3e80ccb6f28c13910ab322406fc7e393ec0dd06d4573c958aadd2b6f3fa62fa0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:24.269441Z","signature_b64":"fSK0xinGiuyuMNVSSumdqzjVSRF7g/Yaq6wY5uBlLtFeFXKoy+DKehuarQZOYjv1XZ/e6L/gFkryt/Qt2qwuAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e80ccb6f28c13910ab322406fc7e393ec0dd06d4573c958aadd2b6f3fa62fa0","last_reissued_at":"2026-07-05T09:16:24.268931Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:24.268931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.03743","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":"2026-07-05T09:16:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xpykU+IEkFqc+FTNplef/tKABKRi61PBSDN0ixsYoCfLpVyJskEKPraeBImUKQA4mLeCPzKRAQa0mZDkbd0wCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T06:01:03.978841Z"},"content_sha256":"796a41017654195e56f6928e5390628eb06ee8b90680dd3e7e06a646ace1e2df","schema_version":"1.0","event_id":"sha256:796a41017654195e56f6928e5390628eb06ee8b90680dd3e7e06a646ace1e2df"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:H2AMZNXSRQJZCCVTEJAG7R7DSP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Mitigating Training Imbalance in LLM Fine-Tuning via Selective Parameter Merging","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Hanyu Zhao, Siqi Fan, Xingrun Xing, Yiming Ju, Zheng Zhang, Zhixiong Zeng, Ziyi Ni","submitted_at":"2024-10-01T08:44:31Z","abstract_excerpt":"Supervised fine-tuning (SFT) is crucial for adapting Large Language Models (LLMs) to specific tasks. In this work, we demonstrate that the order of training data can lead to significant training imbalances, potentially resulting in performance degradation. Consequently, we propose to mitigate this imbalance by merging SFT models fine-tuned with different data orders, thereby enhancing the overall effectiveness of SFT. Additionally, we introduce a novel technique, \"parameter-selection merging,\" which outperforms traditional weighted-average methods on five datasets. Further, through analysis an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03743","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/2410.03743/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":"2026-07-05T09:16:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S+jlgTK9zZ3RdvmwkLN/GN0IH0wEu8P0klIr8hx8h2ApRy57+adgwRgjuuF/2pEIUDHYqaPl5AsgOubIF8ptDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-22T06:01:03.979384Z"},"content_sha256":"6bf17fbebffb4393a47e59c88f3536063e12d5dd90c1a15d1ada5e38cd5d05fb","schema_version":"1.0","event_id":"sha256:6bf17fbebffb4393a47e59c88f3536063e12d5dd90c1a15d1ada5e38cd5d05fb"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/bundle.json","state_url":"https://pith.science/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/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":"2026-08-22T06:01:03Z","links":{"resolver":"https://pith.science/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP","bundle":"https://pith.science/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/bundle.json","state":"https://pith.science/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/H2AMZNXSRQJZCCVTEJAG7R7DSP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:H2AMZNXSRQJZCCVTEJAG7R7DSP","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":"dd72ac23c33bf06abf94c2dfa1233308d6b587a2a8e8014c7affb26d4362313c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-01T08:44:31Z","title_canon_sha256":"796373efc50b7a50220c7f5fb6b68afe2ba58888c28ef2839f4f35f77506155b"},"schema_version":"1.0","source":{"id":"2410.03743","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03743","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03743v1","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03743","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_12","alias_value":"H2AMZNXSRQJZ","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_16","alias_value":"H2AMZNXSRQJZCCVT","created_at":"2026-07-05T09:16:24Z"},{"alias_kind":"pith_short_8","alias_value":"H2AMZNXS","created_at":"2026-07-05T09:16:24Z"}],"graph_snapshots":[{"event_id":"sha256:6bf17fbebffb4393a47e59c88f3536063e12d5dd90c1a15d1ada5e38cd5d05fb","target":"graph","created_at":"2026-07-05T09:16:24Z","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/2410.03743/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Supervised fine-tuning (SFT) is crucial for adapting Large Language Models (LLMs) to specific tasks. In this work, we demonstrate that the order of training data can lead to significant training imbalances, potentially resulting in performance degradation. Consequently, we propose to mitigate this imbalance by merging SFT models fine-tuned with different data orders, thereby enhancing the overall effectiveness of SFT. Additionally, we introduce a novel technique, \"parameter-selection merging,\" which outperforms traditional weighted-average methods on five datasets. Further, through analysis an","authors_text":"Hanyu Zhao, Siqi Fan, Xingrun Xing, Yiming Ju, Zheng Zhang, Zhixiong Zeng, Ziyi Ni","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-01T08:44:31Z","title":"Mitigating Training Imbalance in LLM Fine-Tuning via Selective Parameter Merging"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03743","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":"sha256:796a41017654195e56f6928e5390628eb06ee8b90680dd3e7e06a646ace1e2df","target":"record","created_at":"2026-07-05T09:16:24Z","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":"dd72ac23c33bf06abf94c2dfa1233308d6b587a2a8e8014c7affb26d4362313c","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-01T08:44:31Z","title_canon_sha256":"796373efc50b7a50220c7f5fb6b68afe2ba58888c28ef2839f4f35f77506155b"},"schema_version":"1.0","source":{"id":"2410.03743","kind":"arxiv","version":1}},"canonical_sha256":"3e80ccb6f28c13910ab322406fc7e393ec0dd06d4573c958aadd2b6f3fa62fa0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3e80ccb6f28c13910ab322406fc7e393ec0dd06d4573c958aadd2b6f3fa62fa0","first_computed_at":"2026-07-05T09:16:24.268931Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:16:24.268931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fSK0xinGiuyuMNVSSumdqzjVSRF7g/Yaq6wY5uBlLtFeFXKoy+DKehuarQZOYjv1XZ/e6L/gFkryt/Qt2qwuAw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:16:24.269441Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.03743","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:796a41017654195e56f6928e5390628eb06ee8b90680dd3e7e06a646ace1e2df","sha256:6bf17fbebffb4393a47e59c88f3536063e12d5dd90c1a15d1ada5e38cd5d05fb"],"state_sha256":"c13a014009efc1c128384e9279e8f6fa0b7f5cbc80a8b16c1f4c3030adac1f18"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pMKB+HL7HOwOD9BGJnJW+uH+JWTkPybemiGClgR9BpH823QI+9u+KW/mLxMJfhv5jfXy70EH7b11kcCnBh5TDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-22T06:01:03.984065Z","bundle_sha256":"f0902dd0afa5b0b323df7e8e1aef010d6cc48fb91b0fcffb25394ab23e5e36ba"}}