1
Projector Is All You Train
投影器成训练核心？这篇论文挑战传统微调范式，或为高效训练提供新思路。
arXiv:2608.19726v1 Announce Type: cross Abstract: The typical training process of a multimodal large language model (MLLM) involves adapting both the …
2
Learning to Adapt Cross-Domain Preferences via Meta-LoRA for LLM Personalization
元学习+LoRA让大模型快速适应跨域偏好，个性化调校从此更聪明高效。
arXiv:2608.12389v1 Announce Type: new Abstract: Cross-domain zero- or few-shot personalization aims to generate user-preferred responses in unseen con…
3
Efficiency vs. Alignment: Investigating Safety and Fairness Risks in Parameter-Efficient Fine-Tuning of LLMs
低资源微调大模型可能暗藏安全与公平性隐患，这项研究系统揭示了PEFT的对齐风险。
arXiv:2511.00382v2 Announce Type: replace Abstract: Organizations increasingly adapt Large Language Models (LLMs) from public repositories such as Hug…
4
ReCoLoRA: Spectrum-Aware Recursive Consolidation for Continual LLM Fine-Tuning
突破LLM持续微调遗忘瓶颈，ReCoLoRA以频谱感知递归合并实现任务序列高效学习。
arXiv:2607.07719v1 Announce Type: new Abstract: Parameter-efficient fine-tuning adapts a large language model to one task cheaply, but across a task s…
5
Parameter Efficient Multimodal Instruction Tuning for Romanian Vision Language Models
面向低资源语言罗马尼亚语的多模态指令微调，用参数高效方法实现视觉语言模型适配，填补非英语VLM研究空白。
arXiv:2512.14926v2 Announce Type: replace-cross Abstract: Focusing on low-resource languages is an essential step toward democratizing generative AI. …
6
FedRot-LoRA: Mitigating Rotational Misalignment in Federated LoRA
联邦LoRA新方案FedRot-LoRA，创新解决旋转错位问题，提升模型聚合效率与收敛速度。
arXiv:2602.23638v3 Announce Type: replace-cross Abstract: Federated LoRA provides a communication-efficient mechanism for fine-tuning large language m…
7
Advancing the State-of-the-Art in Empirical Privacy Auditing
新研究前沿：针对大语言模型的参数高效微调，如何通过经验隐私审计量化个体记忆泄露风险，提升数据安全。
arXiv:2606.10481v1 Announce Type: cross Abstract: Parameter-efficient fine-tuning of large language models (LLMs) can exhibit problematic memorization…
8
Instruction Finetuning DeepSeek-R1-8B Model Using LoRA and NEFTune
探索用LoRA和NEFTune方法高效微调DeepSeek-R1-8B，降低资源消耗同时提升性能。
arXiv:2606.10392v1 Announce Type: new Abstract: Financial named-entity recognition (NER) is essential for translating unstructured financial reports a…
9
CoMoL: Efficient Mixture of LoRA Experts via Dynamic Core Space Merging
提出动态核心空间合并方法，大幅降低混合LoRA专家模型的内存占用与计算开销
arXiv:2603.00573v2 Announce Type: replace Abstract: Large language models (LLMs) achieve remarkable performance on diverse downstream and domain-speci…
10
Leave it to the Specialist: Repair Sparse LLMs with Sparse Fine-Tuning via Sparsity Evolution
提出通过稀疏性演化进行稀疏微调，高效修复稀疏大语言模型，平衡性能与计算开销。
arXiv:2505.24037v3 Announce Type: replace Abstract: Sparse large language models (LLMs) offer an attractive direction toward efficient deployment, but…
11
NaRA: Noise-Aware LoRA for Parameter-Efficient Fine-Tuning of Diffusion LLMs
NaRA提出噪声感知LoRA，显著提升扩散大语言模型参数高效微调效果
arXiv:2605.29716v1 Announce Type: new Abstract: Diffusion Large Language Models (dLLMs) have emerged as a promising non-autoregressive generative para…
12
Learning in the Fisher Subspace: A Guided Initialization for LoRA Fine-Tuning
LoRA新思路：在Fisher子空间中引导初始化，提升微调性能与稳定性。
arXiv:2605.01046v3 Announce Type: replace Abstract: LoRA adapts large language models (LLMs) by restricting updates to low-rank subspaces of pre-train…
13
LLM Zeroth-Order Fine-Tuning is an Inference Workload
零阶微调本质是推理负载，颠覆大模型微调的计算认知
arXiv:2605.28760v1 Announce Type: new Abstract: Zeroth-order (ZO) fine-tuning is attractive for large language models because it replaces backpropagat…
14
Convex Optimization for Alignment and Preference Learning on a Single GPU
单GPU实现凸优化方法，高效解决LLM偏好对齐难题，降低RLHF计算成本。
arXiv:2605.23244v1 Announce Type: new Abstract: Fine-tuning large language models (LLMs) to align with human preferences has driven the success of sys…
15
Sparse Orthogonal Parameters Tuning for Continual Learning
arXiv最新论文提出稀疏正交参数调优方法，仅需微调少量正交参数即可有效缓解持续学习中的灾难性遗忘。
arXiv:2411.02813v3 Announce Type: replace Abstract: Continual learning methods based on pre-trained models (PTM) have recently gained attention which …
16
From Parameters to Data: A Task-Parameter-Guided Fine-Tuning Pipeline for Efficient LLM Alignment
提出强映射假设，揭示数据选择与参数高效微调的内在耦合，为LLM对齐提供新思路。
arXiv:2605.21558v1 Announce Type: cross Abstract: Adapting Large Language Models (LLMs) to specialized domains typically incurs high data and computat…
17
Learning Rate Matters: Vanilla LoRA May Suffice for LLM Fine-tuning
通过调整学习率，简单LoRA即可媲美复杂微调方法，揭示被忽视的关键因素。
arXiv:2602.04998v2 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) is the prevailing approach for efficient large language model (LLM) fin…
18
TiTok: Transfer Token-level Knowledge via Contrastive Excess to Transplant LoRA
提出TiTok方法，通过对比学习转移Token级知识，让LoRA参数可在不同骨干模型间移植。
arXiv:2510.04682v3 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are widely applied in real world scenarios, yet fine-tuning the…
19
PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts
多任务微调新方法，用连续提示优化实现参数高效的大模型适配，减少数据需求。
arXiv:2605.14055v1 Announce Type: cross Abstract: Parameter-Efficient Fine-Tuning (PEFT) is widely used for adapting Large Language Models (LLMs) for …
20
Strategic Over-Parameterization for Generalizable Low-Rank Adaptation
策略性过参数化提升低秩适应泛化能力，为参数高效微调提供新理论。
arXiv:2605.16470v1 Announce Type: new Abstract: Adapting large language models (LLMs) to downstream tasks via full fine-tuning is increasingly impract…