LoRA能有效提升模型推理能力，但对规划能力帮助有限。
Planning vs Reasoning: Ablations to Test Capabilities of LoRA layers
- 用低秩适配层微调模型，测试其推理与规划能力差异。
- 推理任务仅需2-3倍低秩空间即可达成效果，规划则需更高秩。
- 适合研究高效微调与模型泛化机制的读者。
低秩适配（LoRA）层已成为高效模型微调的有前景方法，但其能力与局限尚未充分探索。本文：1）探究LoRA层是否能有效增强推理与规划能力；2）提出HashChain Reasoning新评估数据集，可确定性测试推理能力。通过对GPT-2的系统消融实验发现，推理能力主要存在于低秩空间，且可通过LoRA层有效增强。训练后LoRA矩阵的有效秩分析显示，推理任务所需秩仅为规划任务的2-3倍，揭示了LoRA在低参数空间中更适用于推理任务的潜力，也表明推理在泛化上更偏好低秩表示。
原文摘要 · Abstract (English)
Low-Rank Adaptation (LoRA) layers have emerged as a promising approach for efficient model fine-tuning, but their capabilities and limitations have not been fully explored. This paper: 1) Investigates the fundamental question of whether LoRA layers are effective at increasing reasoning + planning abilities 2) We introduce HashChain Reasoning, a novel evaluation dataset that deterministically tests reasoning capabilities. Through systematic ablation studies on GPT-2, we demonstrate that reasoning capabilities appear to exist primarily in low-rank spaces and can be effectively enhanced using LoRA layers. The effective rank analysis of trained LoRA matrices reveals a 2-3x lower rank requirement for reasoning tasks compared to planning tasks, giving context on where LoRA layers would be effective. This also provides evidence for reasoning fundamentally preferring low-parameter spaces for generalization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。