{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QOJJAV5VRSBAO26LE3YBUANG7R","short_pith_number":"pith:QOJJAV5V","schema_version":"1.0","canonical_sha256":"83929057b58c82076bcb26f01a01a6fc4579ae8830be46487380335b731e1fd8","source":{"kind":"arxiv","id":"2404.15159","version":3},"attestation_state":"computed","paper":{"title":"MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cal Yang, Dengchun Li, Jie Zuo, Lei Duan, Mingjie Tang, Naizheng Wang, Yan Zhang, Yinghao Tang, Yingzi Ma, Zhengmao Ye, Zhiyuan Cheng","submitted_at":"2024-04-22T02:15:52Z","abstract_excerpt":"Fine-tuning Large Language Models (LLMs) is a common practice to adapt pre-trained models for specific applications. While methods like LoRA have effectively addressed GPU memory constraints during fine-tuning, their performance often falls short, especially in multi-task scenarios. In contrast, Mixture-of-Expert (MoE) models, such as Mixtral 8x7B, demonstrate remarkable performance in multi-task learning scenarios while maintaining a reduced parameter count. However, the resource requirements of these MoEs remain challenging, particularly for consumer-grade GPUs with less than 24GB memory. To"},"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":"2404.15159","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-22T02:15:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8ac4a13ca7eefbb88a1ef119dbdee4867318e33f59b289987c6099696cbc20e2","abstract_canon_sha256":"ee556ae81d557be3eacb3219f54e7300900feab991d4f1196d3406d24fc9268d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:46:24.418792Z","signature_b64":"zMVJ9Ahiqrd7wAqpPYLmbi2SxxUoIKg/eg+WtQga5Pu6+EICFa4pwFt8f690NHs14Uv/YFQ2ZJcG2Sa23StjAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"83929057b58c82076bcb26f01a01a6fc4579ae8830be46487380335b731e1fd8","last_reissued_at":"2026-07-05T08:46:24.418292Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:46:24.418292Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MixLoRA: Enhancing Large Language Models Fine-Tuning with LoRA-based Mixture of Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cal Yang, Dengchun Li, Jie Zuo, Lei Duan, Mingjie Tang, Naizheng Wang, Yan Zhang, Yinghao Tang, Yingzi Ma, Zhengmao Ye, Zhiyuan Cheng","submitted_at":"2024-04-22T02:15:52Z","abstract_excerpt":"Fine-tuning Large Language Models (LLMs) is a common practice to adapt pre-trained models for specific applications. While methods like LoRA have effectively addressed GPU memory constraints during fine-tuning, their performance often falls short, especially in multi-task scenarios. In contrast, Mixture-of-Expert (MoE) models, such as Mixtral 8x7B, demonstrate remarkable performance in multi-task learning scenarios while maintaining a reduced parameter count. However, the resource requirements of these MoEs remain challenging, particularly for consumer-grade GPUs with less than 24GB memory. To"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.15159","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/2404.15159/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":"2404.15159","created_at":"2026-07-05T08:46:24.418358+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.15159v3","created_at":"2026-07-05T08:46:24.418358+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.15159","created_at":"2026-07-05T08:46:24.418358+00:00"},{"alias_kind":"pith_short_12","alias_value":"QOJJAV5VRSBA","created_at":"2026-07-05T08:46:24.418358+00:00"},{"alias_kind":"pith_short_16","alias_value":"QOJJAV5VRSBAO26L","created_at":"2026-07-05T08:46:24.418358+00:00"},{"alias_kind":"pith_short_8","alias_value":"QOJJAV5V","created_at":"2026-07-05T08:46:24.418358+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":21,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.07023","citing_title":"Online Data Selection Is Implicit Alignment","ref_index":48,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21645","citing_title":"Behavioral and Representational Evidence of Binomial Ordering Preferences in Large Language Models","ref_index":128,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19549","citing_title":"Predicting Mergeability of Parameter-Efficient Fine-Tuning Updates","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31432","citing_title":"Clinically Structured Rank-Gated LoRA for Cross-Benchmark Medical Question Answering","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31432","citing_title":"Clinically Structured Rank-Gated LoRA for Cross-Benchmark Medical Question Answering","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07500","citing_title":"Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03979","citing_title":"Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories","ref_index":79,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31432","citing_title":"Clinically Structured Rank-Gated LoRA for Cross-Benchmark Medical Question Answering","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29375","citing_title":"TriageRA-CCF: Source-Side Clinical Confidence and Coverage Signals for Adaptive Rank Budgeting in Medical LLMs","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29425","citing_title":"Mixture of Debaters: Learn to Debate at Architectural Level in Multi-Agent Reasoning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26376","citing_title":"BioFact-MoE: Biologically Factorized Mixture of Experts for Vision-Language Prognostic Modeling in Hepatocellular Carcinoma","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2410.06431","citing_title":"Functional-level Uncertainty Quantification for Calibrated Fine-tuning on LLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2506.21035","citing_title":"Little by Little: Continual Learning via Incremental Mixture of Rank-1 Associative Memory Experts","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07111","citing_title":"Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2506.01897","citing_title":"MLorc: Momentum Low-rank Compression for Memory Efficient Large Language Model Adaptation","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2507.00029","citing_title":"LoRA-Mixer: Coordinate Modular LoRA Experts Through Serial Attention Routing","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2403.14608","citing_title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","ref_index":96,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26340","citing_title":"Adaptive and Fine-grained Module-wise Expert Pruning for Efficient LoRA-MoE Fine-Tuning","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19254","citing_title":"ShadowPEFT: Shadow Network for Parameter-Efficient Fine-Tuning","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07111","citing_title":"Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06291","citing_title":"TalkLoRA: Communication-Aware Mixture of Low-Rank Adaptation for Large Language Models","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R","json":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R.json","graph_json":"https://pith.science/api/pith-number/QOJJAV5VRSBAO26LE3YBUANG7R/graph.json","events_json":"https://pith.science/api/pith-number/QOJJAV5VRSBAO26LE3YBUANG7R/events.json","paper":"https://pith.science/paper/QOJJAV5V"},"agent_actions":{"view_html":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R","download_json":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R.json","view_paper":"https://pith.science/paper/QOJJAV5V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.15159&json=true","fetch_graph":"https://pith.science/api/pith-number/QOJJAV5VRSBAO26LE3YBUANG7R/graph.json","fetch_events":"https://pith.science/api/pith-number/QOJJAV5VRSBAO26LE3YBUANG7R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R/action/storage_attestation","attest_author":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R/action/author_attestation","sign_citation":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R/action/citation_signature","submit_replication":"https://pith.science/pith/QOJJAV5VRSBAO26LE3YBUANG7R/action/replication_record"}},"created_at":"2026-07-05T08:46:24.418358+00:00","updated_at":"2026-07-05T08:46:24.418358+00:00"}