{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YFVVQ2RYI2FXEN67X44S2XQDMF","short_pith_number":"pith:YFVVQ2RY","schema_version":"1.0","canonical_sha256":"c16b586a38468b7237dfbf392d5e0361747758d81fd4105bb59882cc76e05950","source":{"kind":"arxiv","id":"2606.05613","version":1},"attestation_state":"computed","paper":{"title":"Multilingual Fine-Tuning via Localized Gradient Conflict Resolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Long P. Hoang, Wei Lu, Wenxuan Zhang, Yiran Zhao","submitted_at":"2026-06-04T02:36:30Z","abstract_excerpt":"The rapid evolution of Large Language Models (LLMs) has established cross-lingual versatility as a defining feature of modern systems. However, fine-tuning these models frequently induces negative interference across languages. To address this, we reformulate multilingual fine-tuning as a multi-objective optimization (MOO) problem. Specifically, we introduce Bucket-Level MOO, a scalable distributed framework that applies gradient-based MOO algorithms locally on parameter buckets. This enables conflict-aware updates without the prohibitive communication overhead of reconstructing full gradient "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.05613","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-04T02:36:30Z","cross_cats_sorted":[],"title_canon_sha256":"beb7b48fc0982bb0025ce1ed3957600e0dfff46bd8d274da20f2fdfd413ef445","abstract_canon_sha256":"8a4139011f050c8d3094210cecf7cc2fe1576916cae4c0639fcf124e376d9e78"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T01:14:56.451147Z","signature_b64":"8Rn+1z0i35YkKma1d8jKkctbqwsKXGY+fNQglZX/QdvErbAqSLlaAD/T80sGJPMLGqZWUfvHWezX9lIabuHWCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c16b586a38468b7237dfbf392d5e0361747758d81fd4105bb59882cc76e05950","last_reissued_at":"2026-06-05T01:14:56.450693Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T01:14:56.450693Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multilingual Fine-Tuning via Localized Gradient Conflict Resolution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Long P. Hoang, Wei Lu, Wenxuan Zhang, Yiran Zhao","submitted_at":"2026-06-04T02:36:30Z","abstract_excerpt":"The rapid evolution of Large Language Models (LLMs) has established cross-lingual versatility as a defining feature of modern systems. However, fine-tuning these models frequently induces negative interference across languages. To address this, we reformulate multilingual fine-tuning as a multi-objective optimization (MOO) problem. Specifically, we introduce Bucket-Level MOO, a scalable distributed framework that applies gradient-based MOO algorithms locally on parameter buckets. This enables conflict-aware updates without the prohibitive communication overhead of reconstructing full gradient "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05613","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/2606.05613/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.05613","created_at":"2026-06-05T01:14:56.450770+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05613v1","created_at":"2026-06-05T01:14:56.450770+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05613","created_at":"2026-06-05T01:14:56.450770+00:00"},{"alias_kind":"pith_short_12","alias_value":"YFVVQ2RYI2FX","created_at":"2026-06-05T01:14:56.450770+00:00"},{"alias_kind":"pith_short_16","alias_value":"YFVVQ2RYI2FXEN67","created_at":"2026-06-05T01:14:56.450770+00:00"},{"alias_kind":"pith_short_8","alias_value":"YFVVQ2RY","created_at":"2026-06-05T01:14:56.450770+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF","json":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF.json","graph_json":"https://pith.science/api/pith-number/YFVVQ2RYI2FXEN67X44S2XQDMF/graph.json","events_json":"https://pith.science/api/pith-number/YFVVQ2RYI2FXEN67X44S2XQDMF/events.json","paper":"https://pith.science/paper/YFVVQ2RY"},"agent_actions":{"view_html":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF","download_json":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF.json","view_paper":"https://pith.science/paper/YFVVQ2RY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05613&json=true","fetch_graph":"https://pith.science/api/pith-number/YFVVQ2RYI2FXEN67X44S2XQDMF/graph.json","fetch_events":"https://pith.science/api/pith-number/YFVVQ2RYI2FXEN67X44S2XQDMF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF/action/storage_attestation","attest_author":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF/action/author_attestation","sign_citation":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF/action/citation_signature","submit_replication":"https://pith.science/pith/YFVVQ2RYI2FXEN67X44S2XQDMF/action/replication_record"}},"created_at":"2026-06-05T01:14:56.450770+00:00","updated_at":"2026-06-05T01:14:56.450770+00:00"}