{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CMDTDYMS2IVF5BYBA24RIGDD6P","short_pith_number":"pith:CMDTDYMS","schema_version":"1.0","canonical_sha256":"130731e192d22a5e870106b9141863f3d191dbc23fa213fb7acb642571a84959","source":{"kind":"arxiv","id":"2505.14256","version":1},"attestation_state":"computed","paper":{"title":"FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Li, Deyi Xiong, Shaolin Zhu, Tianyu Dong","submitted_at":"2025-05-20T12:09:17Z","abstract_excerpt":"In this paper, we present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM). We adopt a two-stage strategy to train FuxiMT. We first pre-train the model on a massive Chinese corpus and then conduct multilingual fine-tuning on a large parallel dataset encompassing 65 languages. FuxiMT incorporates Mixture-of-Experts (MoEs) and employs a curriculum learning strategy for robust performance across various resource levels. Experimental results demonstrate that FuxiMT significantly outperforms strong baselines, including state-o"},"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":"2505.14256","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-20T12:09:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"de219a86dce23ca2d37b549c02a78ee5a32a7bf156c8bbe9b3ab254f57468430","abstract_canon_sha256":"17a60b518237a1d028d5c2bfa0fd1401518a9519d9e0cc5228c11e1e74805b35"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:58.511759Z","signature_b64":"Ai5LMRKeCFvdJfT8jS9AT/de3pC6IZIXLgR27qZGzahkhA0bfyunjA/g/DgRMVbnF8nIBczSExrG3emCQQGlBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"130731e192d22a5e870106b9141863f3d191dbc23fa213fb7acb642571a84959","last_reissued_at":"2026-07-05T11:05:58.511337Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:58.511337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"FuxiMT: Sparsifying Large Language Models for Chinese-Centric Multilingual Machine Translation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Li, Deyi Xiong, Shaolin Zhu, Tianyu Dong","submitted_at":"2025-05-20T12:09:17Z","abstract_excerpt":"In this paper, we present FuxiMT, a novel Chinese-centric multilingual machine translation model powered by a sparsified large language model (LLM). We adopt a two-stage strategy to train FuxiMT. We first pre-train the model on a massive Chinese corpus and then conduct multilingual fine-tuning on a large parallel dataset encompassing 65 languages. FuxiMT incorporates Mixture-of-Experts (MoEs) and employs a curriculum learning strategy for robust performance across various resource levels. Experimental results demonstrate that FuxiMT significantly outperforms strong baselines, including state-o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14256","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/2505.14256/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":"2505.14256","created_at":"2026-07-05T11:05:58.511400+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.14256v1","created_at":"2026-07-05T11:05:58.511400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14256","created_at":"2026-07-05T11:05:58.511400+00:00"},{"alias_kind":"pith_short_12","alias_value":"CMDTDYMS2IVF","created_at":"2026-07-05T11:05:58.511400+00:00"},{"alias_kind":"pith_short_16","alias_value":"CMDTDYMS2IVF5BYB","created_at":"2026-07-05T11:05:58.511400+00:00"},{"alias_kind":"pith_short_8","alias_value":"CMDTDYMS","created_at":"2026-07-05T11:05:58.511400+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/CMDTDYMS2IVF5BYBA24RIGDD6P","json":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P.json","graph_json":"https://pith.science/api/pith-number/CMDTDYMS2IVF5BYBA24RIGDD6P/graph.json","events_json":"https://pith.science/api/pith-number/CMDTDYMS2IVF5BYBA24RIGDD6P/events.json","paper":"https://pith.science/paper/CMDTDYMS"},"agent_actions":{"view_html":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P","download_json":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P.json","view_paper":"https://pith.science/paper/CMDTDYMS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.14256&json=true","fetch_graph":"https://pith.science/api/pith-number/CMDTDYMS2IVF5BYBA24RIGDD6P/graph.json","fetch_events":"https://pith.science/api/pith-number/CMDTDYMS2IVF5BYBA24RIGDD6P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P/action/storage_attestation","attest_author":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P/action/author_attestation","sign_citation":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P/action/citation_signature","submit_replication":"https://pith.science/pith/CMDTDYMS2IVF5BYBA24RIGDD6P/action/replication_record"}},"created_at":"2026-07-05T11:05:58.511400+00:00","updated_at":"2026-07-05T11:05:58.511400+00:00"}