nlp-lora-finetune
Query: LoRA fine tuning LLM
Results: 50
Date: 2026-07-07T18:52:51.034Z
1. Federated Sketching LoRA: A Flexible Framework for Heterogeneous Collaborative Fine-Tuning of LLMs
Authors: Wenzhi Fang, Dong-Jun Han, Liangqi Yuan, Seyyedali Hosseinalipour, Christopher G. Brinton
Categories: cs.LG
Published: 2025-01-31
arXiv: 2501.19389v4
Abstract:
Fine-tuning large language models (LLMs) on resource-constrained clients remains a challenging problem. Recent works have fused low-rank adaptation (LoRA) techniques with federated fine-tuning to mitigate challenges associated with client model sizes and data scarcity. Still, the heterogeneity of resources remains a critical bottleneck: while higher-rank modules generally enhance performance, varying client capabilities constrain LoRA’s feasible rank range. Existing approaches attempting to resolve this issue either lack analytical justification or impose additional computational overhead, leaving a wide gap for efficient and theoretically-grounded solutions. To address these challenges, we propose federated sketching LoRA (FSLoRA), which leverages a sketching mechanism to enable clients to selectively update submatrices of global LoRA modules maintained by the server. By adjusting the sketching ratios, which determine the ranks of the submatrices on the clients, FSLoRA flexibly adapts to client-specific communication and computational constraints. We provide a rigorous convergence analysis of FSLoRA that characterizes how the sketching ratios affect the convergence rate. Through extensive experiments, we demonstrate that FSLoRA outperforms baselines and significantly improves training efficiency while preserving stable convergence.
2. LoRA-FAIR: Federated LoRA Fine-Tuning with Aggregation and Initialization Refinement
Authors: Jieming Bian, Lei Wang, Letian Zhang, Jie Xu
Categories: cs.LG, cs.CV
Published: 2024-11-22
arXiv: 2411.14961v3
Abstract:
Foundation models (FMs) achieve strong performance across diverse tasks with task-specific fine-tuning, yet full parameter fine-tuning is often computationally prohibitive for large models. Parameter-efficient fine-tuning (PEFT) methods like Low-Rank Adaptation (LoRA) reduce this cost by introducing low-rank matrices for tuning fewer parameters. While LoRA allows for efficient fine-tuning, it requires significant data for adaptation, making Federated Learning (FL) an appealing solution due to its privacy-preserving collaborative framework. However, combining LoRA with FL introduces two key challenges: the \textbf{Server-Side Aggregation Bias}, where server-side averaging of LoRA matrices diverges from the ideal global update, and the \textbf{Client-Side Initialization Lag}, emphasizing the need for consistent initialization across rounds. Existing approaches address these challenges individually, limiting their effectiveness. We propose LoRA-FAIR, a novel method that tackles both issues by introducing a correction term on the server, enhancing aggregation efficiency and accuracy. LoRA-FAIR maintains computational and communication efficiency, yielding superior performance over state-of-the-art methods. Experimental results on ViT and MLP-Mixer models across large-scale datasets demonstrate that LoRA-FAIR consistently achieves performance improvements in FL settings.
3. Learning Rate Matters: Vanilla LoRA May Suffice for LLM Fine-tuning
Authors: Yu-Ang Lee, Ching-Yun Ko, Pin-Yu Chen, Mi-Yen Yeh
Categories: cs.LG, cs.AI, cs.CL
Published: 2026-02-04
arXiv: 2602.04998v2
Abstract:
Low-Rank Adaptation (LoRA) is the prevailing approach for efficient large language model (LLM) fine-tuning. Building on this paradigm, recent studies have proposed alternative initialization strategies, architectural modifications, and optimization adjustments, reporting substantial improvements over vanilla LoRA. However, these gains are often demonstrated under fixed or narrowly tuned hyperparameter settings, despite the known sensitivity of neural networks to training configurations. In this work, we systematically re-evaluate nine representative LoRA variants alongside vanilla LoRA through extensive hyperparameter searches over learning rate, batch size, rank, and training duration. Across tasks spanning mathematical reasoning, commonsense reasoning, code generation, and instruction following at diverse model scales, we find that different LoRA methods favor distinct learning rate ranges. Crucially, once learning rates are properly tuned, all methods achieve similar peak performance (within 1-2%), with only subtle rank-dependent behaviors. These results suggest that vanilla LoRA remains a competitive baseline and that improvements reported under a single training configuration may not reflect consistent methodological advantages. Finally, a second-order analysis attributes the differing optimal learning rate ranges to variations in the largest Hessian eigenvalue, aligning with classical learning theories.
4. Layer-wise LoRA fine-tuning: a similarity metric approach
Authors: Keith Ando Ogawa, Bruno Lopes Yamamoto, Lucas Lauton de Alcantara, Lucas Pellicer, Rosimeire Pereira Costa, Edson Bollis, Anna Helena Reali Costa, Artur Jordao
Categories: cs.LG
Published: 2026-02-05
arXiv: 2602.05988v1
Abstract:
Pre-training Large Language Models (LLMs) on web-scale datasets becomes fundamental for advancing general-purpose AI. In contrast, enhancing their predictive performance on downstream tasks typically involves adapting their knowledge through fine-tuning. Parameter-efficient fine-tuning techniques, such as Low-Rank Adaptation (LoRA), aim to reduce the computational cost of this process by freezing the pre-trained model and updating a smaller number of parameters. In comparison to full fine-tuning, these methods achieve over 99% reduction in trainable parameter count, depending on the configuration. Unfortunately, such a reduction may prove insufficient as LLMs continue to grow in scale. In this work, we address the previous problem by systematically selecting only a few layers to fine-tune using LoRA or its variants. We argue that not all layers contribute equally to the model adaptation. Leveraging this, we identify the most relevant layers to fine-tune by measuring their contribution to changes in internal representations. Our method is orthogonal to and readily compatible with existing low-rank adaptation techniques. We reduce the trainable parameters in LoRA-based techniques by up to 50%, while maintaining the predictive performance across different models and tasks. Specifically, on encoder-only architectures, this reduction in trainable parameters leads to a negligible predictive performance drop on the GLUE benchmark. On decoder-only architectures, we achieve a small drop or even improvements in the predictive performance on mathematical problem-solving capabilities and coding tasks. Finally, this effectiveness extends to multimodal models, for which we also observe competitive results relative to fine-tuning with LoRA modules in all layers. Code is available at: https://github.com/c2d-usp/Layer-wise-LoRA-with-CKA
5. Fine-Tuning Open Video Generators for Cinematic Scene Synthesis: A Small-Data Pipeline with LoRA and Wan2.1 I2V
Authors: Meftun Akarsu, Kerem Catay, Sedat Bin Vedat, Enes Kutay Yarkan, Ilke Senturk, Arda Sar, Dafne Eksioglu
Categories: cs.CV, cs.AI
Published: 2025-10-31
arXiv: 2510.27364v1
Abstract:
We present a practical pipeline for fine-tuning open-source video diffusion transformers to synthesize cinematic scenes for television and film production from small datasets. The proposed two-stage process decouples visual style learning from motion generation. In the first stage, Low-Rank Adaptation (LoRA) modules are integrated into the cross-attention layers of the Wan2.1 I2V-14B model to adapt its visual representations using a compact dataset of short clips from Ay Yapim’s historical television film El Turco. This enables efficient domain transfer within hours on a single GPU. In the second stage, the fine-tuned model produces stylistically consistent keyframes that preserve costume, lighting, and color grading, which are then temporally expanded into coherent 720p sequences through the model’s video decoder. We further apply lightweight parallelization and sequence partitioning strategies to accelerate inference without quality degradation. Quantitative and qualitative evaluations using FVD, CLIP-SIM, and LPIPS metrics, supported by a small expert user study, demonstrate measurable improvements in cinematic fidelity and temporal stability over the base model. The complete training and inference pipeline is released to support reproducibility and adaptation across cinematic domains.
6. LoRA-C: Parameter-Efficient Fine-Tuning of Robust CNN for IoT Devices
Authors: Chuntao Ding, Xu Cao, Jianhang Xie, Linlin Fan, Shangguang Wang, Zhichao Lu
Categories: cs.DC
Published: 2024-10-22
arXiv: 2410.16954v2
Abstract:
Efficient fine-tuning of pre-trained convolutional neural network (CNN) models using local data is essential for providing high-quality services to users using ubiquitous and resource-limited Internet of Things (IoT) devices. Low-Rank Adaptation (LoRA) fine-tuning has attracted widespread attention from industry and academia because it is simple, efficient, and does not incur any additional reasoning burden. However, most of the existing advanced methods use LoRA to fine-tune Transformer, and there are few studies on using LoRA to fine-tune CNN. The CNN model is widely deployed on IoT devices for application due to its advantages in comprehensive resource occupancy and performance. Moreover, IoT devices are widely deployed outdoors and usually process data affected by the environment (such as fog, snow, rain, etc.). The goal of this paper is to use LoRA technology to efficiently improve the robustness of the CNN model. To this end, this paper first proposes a strong, robust CNN fine-tuning method for IoT devices, LoRA-C, which performs low-rank decomposition in convolutional layers rather than kernel units to reduce the number of fine-tuning parameters. Then, this paper analyzes two different rank settings in detail and observes that the best performance is usually achieved when $α/{r}$ is a constant in either standard data or corrupted data. This discovery provides experience for the widespread application of LoRA-C. Finally, this paper conducts many experiments based on pre-trained models. Experimental results on CIFAR-10, CIFAR-100, CIFAR-10-C, and Icons50 datasets show that the proposed LoRA-Cs outperforms standard ResNets. Specifically, on the CIFAR-10-C dataset, the accuracy of LoRA-C-ResNet-101 achieves 83.44% accuracy, surpassing the standard ResNet-101 result by +9.5%.
7. Targeted Lexical Injection: Unlocking Latent Cross-Lingual Alignment in Lugha-Llama via Early-Layer LoRA Fine-Tuning
Authors: Stanley Ngugi
Categories: cs.CL
Published: 2025-06-18
arXiv: 2506.15415v1
Abstract:
Large Language Models (LLMs) have demonstrated remarkable capabilities, yet their performance in low-resource languages (LRLs), such as Swahili, often lags due to data scarcity and underrepresentation in pre-training. A key challenge is achieving robust cross-lingual lexical alignment, crucial for tasks like translation and cross-lingual information retrieval. This paper introduces Targeted Lexical Injection (TLI), a novel and efficient fine-tuning approach. We first demonstrate that Lugha-Llama-8B-wura, a Swahili-centric LLM, exhibits strong, near-perfect lexical alignment for Swahili-English word pairs in its early internal layers (specifically Layer 2, with ~0.99998 average cosine similarity based on a pilot study), a capability not fully reflected in its final output representations (baseline ~0.32 similarity on our evaluation set). TLI leverages this insight by using Low-Rank Adaptation (LoRA) and a contrastive learning objective to fine-tune the model, specifically targeting embeddings from this empirically identified optimal early layer. Our experiments show that TLI significantly improves the output-level lexical alignment for 623 trained Swahili-English word pairs, increasing average cosine similarity from 0.3211 to 0.4113 (+28.08%, p < 1.33 x 10^-240). More importantly, these improvements generalize remarkably well to 63 unseen control word pairs, with similarity increasing from 0.3143 to 0.4033 (+28.32%, p < 7.17 x 10^-27). These findings suggest TLI enhances the model’s ability to preserve and propagate its inherent early-layer cross-lingual knowledge, offering a parameter-efficient and effective strategy for improving lexical alignment in LRL-focused LLMs.
8. Is The Universal Matter - Antimatter Asymmetry Fine Tuned?
Authors: Gary Steigman, Robert J. Scherrer
Categories: astro-ph.CO, hep-ph
Published: 2018-01-30
arXiv: 1801.10059v1
Abstract:
The asymmetry between matter and antimatter is key to the existence and nature of our Universe. A measure of the matter - antimatter asymmetry of the Universe is provided by the present value of the universal ratio of baryons (baryons minus antibaryons) to photons (or the ratio of baryons to entropy). The baryon asymmetry parameter is an important physical and cosmological parameter. But how fine tuned is it? A “natural” value for this parameter is zero, corresponding to equal amounts of matter and antimatter. Another, also possibly natural, choice for this dimensionless parameter would be of order unity, corresponding to nearly equal amounts (by number) of matter (and essentially no antimatter) and photons in every comoving volume. However, observations suggest that in the Universe we inhabit the value of this parameter is nonzero, but smaller than this natural value by some nine to ten orders of magnitude. In this contribution we review the evidence that our Universe does not contain equal amounts of matter and antimatter. Any change in the magnitude of the baryon asymmetry parameter necessarily leads to a universe with physical characteristics different from those in our own. The degree of fine tuning in the baryon asymmetry parameter is determined by the width of the range over which it can be varied and still allow for the existence of life. Our results suggest that the baryon asymmetry parameter can be varied over a very wide range without impacting the prospects for life; this result is not suggestive of fine tuning. [abridged]
9. LoRA Land: 310 Fine-tuned LLMs that Rival GPT-4, A Technical Report
Authors: Justin Zhao, Timothy Wang, Wael Abid, Geoffrey Angus, Arnav Garg, Jeffery Kinnison, Alex Sherstinsky, Piero Molino, Travis Addair, Devvret Rishi
Categories: cs.CL, cs.AI, cs.LG
Published: 2024-04-29
arXiv: 2405.00732v1
Abstract:
Low Rank Adaptation (LoRA) has emerged as one of the most widely adopted methods for Parameter Efficient Fine-Tuning (PEFT) of Large Language Models (LLMs). LoRA reduces the number of trainable parameters and memory usage while achieving comparable performance to full fine-tuning. We aim to assess the viability of training and serving LLMs fine-tuned with LoRA in real-world applications. First, we measure the quality of LLMs fine-tuned with quantized low rank adapters across 10 base models and 31 tasks for a total of 310 models. We find that 4-bit LoRA fine-tuned models outperform base models by 34 points and GPT-4 by 10 points on average. Second, we investigate the most effective base models for fine-tuning and assess the correlative and predictive capacities of task complexity heuristics in forecasting the outcomes of fine-tuning. Finally, we evaluate the latency and concurrency capabilities of LoRAX, an open-source Multi-LoRA inference server that facilitates the deployment of multiple LoRA fine-tuned models on a single GPU using shared base model weights and dynamic adapter loading. LoRAX powers LoRA Land, a web application that hosts 25 LoRA fine-tuned Mistral-7B LLMs on a single NVIDIA A100 GPU with 80GB memory. LoRA Land highlights the quality and cost-effectiveness of employing multiple specialized LLMs over a single, general-purpose LLM.
10. YouZhi: Towards High-Concurrency Financial LLMs via Adaptive GQA-to-MLA Transition
Authors: PSBC LLM Team, Huawei LLM Team, Ruihan Long, Junjie Wu, Tianan Zhang, Duo Zhang, Yaozong Wu, Jinbin Fu, Chang Liu, Zhentao Tang, Wenshuang Yang, Xin Wang, Zhihao Song, Ning Huang, Wenjing Xu, Shuai Zong, Shupei Sun, Sen Wang, Jing Hu, Bin Wang, Xinyu Wang, Junkui Ju, Zequn Ding, Jie Ran, Man Luo, Shixiong Kai, Linkai Hou, Kaichao Liang, Hu Zhao, Yang Zhao, Shucheng Lin, Wei Yu, Chenghan Jiang, Jingjing Ding, Jiahui Zhang, Tian Jin, Yuhang Zhang, Dong Guo, Wei Sun, Jun Xie, Jianwei Li, Lei Cao, Pei Li, Jiabin Li, Jia Yuan, Rui Yuan, Jing Zhu, Mingxuan Yuan, Zhangcheng Lv, Xin Jiang, Xiuhong Fei, Xiaozhe Ren, Yulong Li, Zhipeng Zhang, Hang Wang, Zhaohui Xu, Rui Zhao, Yibo He, Xinzhuang Niu
Categories: cs.CL
Published: 2026-06-04
arXiv: 2606.05868v1
Abstract:
Large language models (LLMs) drive significant financial innovations, yet their high-concurrency deployment is severely bottlenecked by KV cache memory overhead, which inflates infrastructure costs and throttles scalability. To address this, we propose YouZhi-LLM, a highly efficient financial LLM empowered by a comprehensive structural transition and training pipeline natively built on the Huawei Ascend ecosystem. At its algorithmic core, YouZhi-LLM features a layer-adaptive GQA-to-MLA transition framework that dynamically assigns per-layer FreqFold sizes, maximizing KV-cache compression while minimizing perplexity degradation. To recover representation capacity and inject domain expertise, the Ascend-based training pipeline seamlessly integrates generalized knowledge distillation with financial-specific supervised fine-tuning. Evaluations demonstrate the superiority of this systematic approach, with the adaptive transition reducing perplexity degradation by up to 35% over uniform baselines. Crucially, when evaluated on Ascend NPUs via vLLM-Ascend, the massive KV-cache reduction translates directly into deployment efficiency. Compared to their respective base models, YouZhi-7B yields a 12.3% improvement in average financial benchmark score alongside a 2.69$\times$ increase in maximum concurrency; similarly, YouZhi-14B achieves a 7.0% accuracy gain and a 2.43$\times$ concurrency boost, establishing a new paradigm for cost-effective, high-throughput financial inference.
11. LoRAFusion: Efficient LoRA Fine-Tuning for LLMs
Authors: Zhanda Zhu, Qidong Su, Yaoyao Ding, Kevin Song, Shang Wang, Gennady Pekhimenko
Categories: cs.LG, cs.AI, cs.DC
Published: 2025-09-30
arXiv: 2510.00206v1
Abstract:
Low-Rank Adaptation (LoRA) has become the leading Parameter-Efficient Fine-Tuning (PEFT) method for Large Language Models (LLMs), as it significantly reduces GPU memory usage while maintaining competitive fine-tuned model quality on downstream tasks. Despite these benefits, we identify two key inefficiencies in existing LoRA fine-tuning systems. First, they incur substantial runtime overhead due to redundant memory accesses on large activation tensors. Second, they miss the opportunity to concurrently fine-tune multiple independent LoRA adapters that share the same base model on the same set of GPUs. This leads to missed performance gains such as reduced pipeline bubbles, better communication overlap, and improved GPU load balance. To address these issues, we introduce LoRAFusion, an efficient LoRA fine-tuning system for LLMs. At the kernel level, we propose a graph-splitting method that fuses memory-bound operations. This design eliminates unnecessary memory accesses and preserves the performance of compute-bound GEMMs without incurring the cost of recomputation or synchronization. At the scheduling level, LoRAFusion introduces an adaptive batching algorithm for multi-job fine-tuning. It first splits LoRA adapters into groups to intentionally stagger batch execution across jobs, and then solves a bin-packing problem within each group to generate balanced, dependency-aware microbatches. LoRAFusion achieves up to $1.96\times$ ($1.47\times$ on average) end-to-end speedup compared to Megatron-LM, and up to $1.46\times$ ($1.29\times$ on average) improvement over mLoRA, the state-of-the-art multi-LoRA fine-tuning system. Our fused kernel achieves up to $1.39\times$ ($1.27\times$ on average) kernel performance improvement and can directly serve as a plug-and-play replacement in existing LoRA systems. We open-source LoRAFusion at https://github.com/CentML/lorafusion.
12. Leaner Training, Lower Leakage: Revisiting Memorization in LLM Fine-Tuning with LoRA
Authors: Fei Wang, Baochun Li
Categories: cs.LG, cs.CL, cs.CR
Published: 2025-06-25
arXiv: 2506.20856v1
Abstract:
Memorization in large language models (LLMs) makes them vulnerable to data extraction attacks. While pre-training memorization has been extensively studied, fewer works have explored its impact in fine-tuning, particularly for LoRA fine-tuning, a widely adopted parameter-efficient method. In this work, we re-examine memorization in fine-tuning and uncover a surprising divergence from prior findings across different fine-tuning strategies. Factors such as model scale and data duplication, which strongly influence memorization in pre-training and full fine-tuning, do not follow the same trend in LoRA fine-tuning. Using a more relaxed similarity-based memorization metric, we demonstrate that LoRA significantly reduces memorization risks compared to full fine-tuning, while still maintaining strong task performance.
13. LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization
Authors: Jui-Nan Yen, Si Si, Zhao Meng, Felix Yu, Sai Surya Duvvuri, Inderjit S. Dhillon, Cho-Jui Hsieh, Sanjiv Kumar
Categories: cs.LG, cs.AI, cs.CL
Published: 2024-10-27
arXiv: 2410.20625v2
Abstract:
Low-rank adaption (LoRA) is a widely used parameter-efficient finetuning method for LLM that reduces memory requirements. However, current LoRA optimizers lack transformation invariance, meaning the actual updates to the weights depends on how the two LoRA factors are scaled or rotated. This deficiency leads to inefficient learning and sub-optimal solutions in practice. This paper introduces LoRA-RITE, a novel adaptive matrix preconditioning method for LoRA optimization, which can achieve transformation invariance and remain computationally efficient. We provide theoretical analysis to demonstrate the benefit of our method and conduct experiments on various LLM tasks with different models including Gemma 2B, 7B, and mT5-XXL. The results demonstrate consistent improvements against existing optimizers. For example, replacing Adam with LoRA-RITE during LoRA fine-tuning of Gemma-2B yielded 4.6% accuracy gain on Super-Natural Instructions and 3.5% accuracy gain across other four LLM benchmarks (HellaSwag, ArcChallenge, GSM8K, OpenBookQA).
14. CLoQ: Enhancing Fine-Tuning of Quantized LLMs via Calibrated LoRA Initialization
Authors: Yanxia Deng, Aozhong Zhang, Selcuk Gurses, Naigang Wang, Zi Yang, Penghang Yin
Categories: cs.LG, cs.AI
Published: 2025-01-30
arXiv: 2501.18475v2
Abstract:
Fine-tuning large language models (LLMs) using low-rank adaptation (LoRA) has become a highly efficient approach for downstream tasks, particularly in scenarios with limited computational resources. However, applying LoRA techniques to quantized LLMs poses unique challenges due to the reduced representational precision of quantized weights. In this paper, we introduce CLoQ (Calibrated LoRA initialization for Quantized LLMs), a simplistic initialization strategy designed to overcome these challenges. Our approach focuses on minimizing the layer-wise discrepancy between the original LLM and its quantized counterpart with LoRA components during initialization. By leveraging a small calibration dataset, CLoQ quantizes a pre-trained LLM and determines the optimal LoRA components for each layer, ensuring a strong foundation for subsequent fine-tuning. A key contribution of this work is a novel theoretical result that enables the accurate and closed-form construction of these optimal LoRA components. We validate the efficacy of CLoQ across multiple tasks such as language generation, arithmetic reasoning, and commonsense reasoning, demonstrating that it consistently outperforms existing LoRA fine-tuning methods for quantized LLMs, especially at ultra low-bit widths.
15. Text to Trust: Evaluating Fine-Tuning and LoRA Trade-offs in Language Models for Unfair Terms of Service Detection
Authors: Noshitha Padma Pratyusha Juttu, Sahithi Singireddy, Sravani Gona, Sujal Timilsina
Categories: cs.CL, cs.AI, cs.LG
Published: 2025-10-26
arXiv: 2510.22531v1
Abstract:
Large Language Models (LLMs) have transformed text understanding, yet their adaptation to specialized legal domains remains constrained by the cost of full fine-tuning. This study provides a systematic evaluation of fine tuning, parameter efficient adaptation (LoRA, QLoRA), and zero-shot prompting strategies for unfair clause detection in Terms of Service (ToS) documents, a key application in legal NLP. We finetune BERT and DistilBERT, apply 4-bit Low-Rank Adaptation (LoRA) to models such as TinyLlama, LLaMA 3B/7B, and SaulLM, and evaluate GPT-4o and O-versions in zero-shot settings. Experiments on the CLAUDETTE-ToS benchmark and the Multilingual Scraper Corpus show that full fine-tuning achieves the strongest precision recall balance, while LoRA-based models provide competitive recall with up to 3x lower memory cost. These findings highlight practical design trade-offs for efficient and domain-adapted LLMs, contributing open baselines for fine-tuning research in legal text processing.
16. Topic Modeling with Fine-tuning LLMs and Bag of Sentences
Authors: Johannes Schneider
Categories: cs.CL, cs.LG
Published: 2024-08-06
arXiv: 2408.03099v2
Abstract:
Large language models (LLMs) are increasingly used for topic modeling, outperforming classical topic models such as LDA. Commonly, pre-trained LLM encoders such as BERT are used out-of-the-box despite the fact that fine-tuning is known to improve LLMs considerably. The challenge lies in obtaining a suitable labeled dataset for fine-tuning. In this paper, we build on the recent idea of using bags of sentences as the elementary unit for computing topics. Based on this idea, we derive an approach called FT-Topic to perform unsupervised fine-tuning, relying primarily on two steps for constructing a training dataset in an automatic fashion. First, a heuristic method identifies pairs of sentence groups that are assumed to belong either to the same topic or to different topics. Second, we remove sentence pairs that are likely labeled incorrectly. The resulting dataset is then used to fine-tune an encoder LLM, which can be leveraged by any topic modeling approach that uses embeddings. In this work, we demonstrate its effectiveness by deriving a novel state-of-the-art topic modeling method called SenClu. The method achieves fast inference through an expectation-maximization algorithm and hard assignments of sentence groups to a single topic, while allowing users to encode prior knowledge about the topic-document distribution. Code is available at https://github.com/JohnTailor/FT-Topic
17. Federated LoRA Fine-Tuning for LLMs via Collaborative Alignment
Authors: Shuaida He, Liwen Chen, Long Feng
Categories: stat.ML, cs.LG
Published: 2026-05-20
arXiv: 2605.21217v1
Abstract:
Low-rank adaptation (LoRA) has emerged as a powerful tool for parameter-efficient fine-tuning of large language models (LLMs). This paper studies LoRA under a federated learning setting, enabling collaborative fine-tuning across clients while preserving parameter efficiency. We focus on a highly heterogeneous regime in which clients share only partial structure and a substantial subset may be contaminated. We propose Collaborative Low-rank Alignment and Identifiable Recovery (CLAIR), a contamination-aware framework that relies only on preliminary local estimators. Its formulation applies broadly, from linear regression to neural network and LLM modules, whenever local adaptation can be represented by matrix-valued updates. CLAIR recovers the shared LoRA subspace and detects contaminated clients via a structured low-rank plus block-sparse decomposition. We prove exact recovery of the shared LoRA subspace in the noiseless case, stable recovery under preliminary estimation error, and consistent collaborative-set recovery under mild separation conditions. We further quantify the gain from CLAIR refinement: it reduces off-subspace estimation error through cross-client averaging while preserving client-specific variation within the shared LoRA subspace, thus improves over local fine-tuning whenever this oracle gain outweighs the costs of subspace estimation and benign-client heterogeneity. Empirically, we demonstrate the benefits of CLAIR by fine-tuning a Transformer architecture on a text-copying task. The results show accurate contamination detection and improved benign-client performance compared with local fine-tuning and non-robust federated averaging.
18. How Relevance Emerges: Interpreting LoRA Fine-Tuning in Reranking LLMs
Authors: Atharva Nijasure, Tanya Chowdhury, James Allan
Categories: cs.IR, cs.LG
Published: 2025-04-05
arXiv: 2504.08780v3
Abstract:
We conduct a behavioral exploration of LoRA fine-tuned LLMs for Passage Reranking to understand how relevance signals are learned and deployed by Large Language Models. By fine-tuning Mistral-7B, LLaMA3.1-8B, and Pythia-6.9B on MS MARCO under diverse LoRA configurations, we investigate how relevance modeling evolves across checkpoints, the impact of LoRA rank (1, 2, 8, 32), and the relative importance of updated MHA vs. MLP components. Our ablations reveal which layers and projections within LoRA transformations are most critical for reranking accuracy. These findings offer fresh explanations into LoRA’s adaptation mechanisms, setting the stage for deeper mechanistic studies in Information Retrieval. All models used in this study have been shared.
19. LoRA-BAM: Input Filtering for Fine-tuned LLMs via Boxed Abstraction Monitors over LoRA Layers
Authors: Changshun Wu, Tianyi Duan, Saddek Bensalem, Chih-Hong Cheng
Categories: cs.LG
Published: 2025-06-01
arXiv: 2506.00998v1
Abstract:
Fine-tuning large language models (LLMs) improves performance on domain-specific tasks but can lead to overfitting, making them unreliable on out-of-distribution (OoD) queries. We propose LoRA-BAM - a method that adds OoD detection monitors to the LoRA layer using boxed abstraction to filter questions beyond the model’s competence. Feature vectors from the fine-tuning data are extracted via the LLM and clustered. Clusters are enclosed in boxes; a question is flagged as OoD if its feature vector falls outside all boxes. To improve interpretability and robustness, we introduce a regularization loss during fine-tuning that encourages paraphrased questions to stay close in the feature space, and the enlargement of the decision boundary is based on the feature variance within a cluster. Our method complements existing defenses by providing lightweight and interpretable OoD detection.
20. Loquetier: A Virtualized Multi-LoRA Framework for Unified LLM Fine-tuning and Serving
Authors: Yuchen Zhang, Hanyue Du, Chun Cao, Jingwei Xu
Categories: cs.LG, cs.AI
Published: 2025-10-30
arXiv: 2511.00101v1
Abstract:
Low-Rank Adaptation (LoRA) has become a widely adopted parameter-efficient fine-tuning (PEFT) technique for adapting large language models (LLMs) to downstream tasks. While prior work has explored strategies for integrating LLM training and serving, there still remains a gap in unifying fine-tuning and inference for LoRA-based models. We present Loquetier, a virtualized multi-LoRA framework that seamlessly integrates LoRA fine-tuning and serving within a single runtime. Loquetier introduces two key components: (1) a Virtualized Module that isolates PEFT-based modifications and supports multiple adapters on a shared base model, and (2) an optimized computation flow with a kernel design that merges fine-tuning and inference paths in forward propagation, enabling efficient batching and minimizing kernel invocation overhead. Extensive experiments across three task settings show that Loquetier consistently outperforms existing baselines in both performance and flexibility, achieving up to $3.0\times$ the throughput of the state-of-the-art co-serving system on inference-only tasks and $46.4\times$ higher SLO attainment than PEFT on unified fine-tuning and inference tasks. The implementation of Loquetier is publicly available at https://github.com/NJUDeepEngine/Loquetier.
21. When Fine-Tuning Fails and when it Generalises: Role of Data Diversity and Mixed Training in LLM-based TTS
Authors: Anupam Purwar, Aditya Choudhary
Categories: cs.SD, cs.AI, cs.ET
Published: 2026-03-11
arXiv: 2603.10904v1
Abstract:
Large language models are increasingly adopted as semantic backbones for neural text-to-speech systems. However, frozen LLM representations are insufficient for modeling speaker specific acoustic and perceptual characteristics. Our experiments involving fine tuning of the Language Model backbone of TTS show promise in improving the voice consistency and Signal to Noise ratio SNR in voice cloning task. Across multiple speakers LoRA finetuning consistently outperforms the non-finetuned base Qwen-0.5B model across three complementary dimensions of speech quality. First, perceptual quality improves significantly with DNS-MOS gains of up to 0.42 points for speakers whose training data exhibits sufficient acoustic variability. Second, speaker fidelity improves for all evaluated speakers with consistent increases in voice similarity indicating that LoRA effectively adapts speaker identity representations without degrading linguistic modeling. Third, signal level quality improves in most cases with signal to noise ratio increasing by as much as 34 percent. Crucially these improvements are strongly governed by the characteristics of the training data. Speakers with high variability in acoustic energy and perceptual quality achieve simultaneous gains in DNS-MOS voice similarity and SNR. Overall this work establishes that LoRA finetuning is not merely a parameter efficient optimization technique but an effective mechanism for better speaker level adaptation in compact LLM-based TTS systems. When supported by sufficiently diverse training data LoRA adapted Qwen-0.5B consistently surpasses its frozen base model in perceptual quality speaker similarity with low latency using GGUF model hosted in quantized form.
22. LowRA: Accurate and Efficient LoRA Fine-Tuning of LLMs under 2 Bits
Authors: Zikai Zhou, Qizheng Zhang, Hermann Kumbong, Kunle Olukotun
Categories: cs.LG, cs.AR, cs.CL, cs.PF
Published: 2025-02-12
arXiv: 2502.08141v1
Abstract:
Fine-tuning large language models (LLMs) is increasingly costly as models scale to hundreds of billions of parameters, and even parameter-efficient fine-tuning (PEFT) methods like LoRA remain resource-intensive. We introduce LowRA, the first framework to enable LoRA fine-tuning below 2 bits per parameter with minimal performance loss. LowRA optimizes fine-grained quantization - mapping, threshold selection, and precision assignment - while leveraging efficient CUDA kernels for scalable deployment. Extensive evaluations across 4 LLMs and 4 datasets show that LowRA achieves a superior performance-precision trade-off above 2 bits and remains accurate down to 1.15 bits, reducing memory usage by up to 50%. Our results highlight the potential of ultra-low-bit LoRA fine-tuning for resource-constrained environments.
23. Evaluating Fine-Tuning Efficiency of Human-Inspired Learning Strategies in Medical Question Answering
Authors: Yushi Yang, Andrew M. Bean, Robert McCraith, Adam Mahdi
Categories: cs.CL
Published: 2024-08-15
arXiv: 2408.07888v2
Abstract:
Fine-tuning Large Language Models (LLMs) incurs considerable training costs, driving the need for data-efficient training with optimised data ordering. Human-inspired strategies offer a solution by organising data based on human learning practices. This study evaluates the fine-tuning efficiency of five human-inspired strategies across four language models, three datasets, and both human- and LLM-labelled data in the context of medical question answering. These strategies achieve the best accuracy gain of 1.81% and an average gain of 1.02% across datasets, with interleaved strategies delivering the best average results. However, the best strategy varies across model-dataset combinations, limiting the generalisability of the effects of any single strategy. Additionally, LLM-defined question difficulty outperforms human-defined labels in curriculum-based learning, showing the potential of model-generated data as a cost-effective alternative for optimising fine-tuning.
24. FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learning
Authors: Weirui Kuang, Bingchen Qian, Zitao Li, Daoyuan Chen, Dawei Gao, Xuchen Pan, Yuexiang Xie, Yaliang Li, Bolin Ding, Jingren Zhou
Categories: cs.LG
Published: 2023-09-01
arXiv: 2309.00363v1
Abstract:
LLMs have demonstrated great capabilities in various NLP tasks. Different entities can further improve the performance of those LLMs on their specific downstream tasks by fine-tuning LLMs. When several entities have similar interested tasks, but their data cannot be shared because of privacy concerns regulations, federated learning (FL) is a mainstream solution to leverage the data of different entities. However, fine-tuning LLMs in federated learning settings still lacks adequate support from existing FL frameworks because it has to deal with optimizing the consumption of significant communication and computational resources, data preparation for different tasks, and distinct information protection demands. This paper first discusses these challenges of federated fine-tuning LLMs, and introduces our package FS-LLM as a main contribution, which consists of the following components: (1) we build an end-to-end benchmarking pipeline, automizing the processes of dataset preprocessing, federated fine-tuning execution, and performance evaluation on federated LLM fine-tuning; (2) we provide comprehensive federated parameter-efficient fine-tuning algorithm implementations and versatile programming interfaces for future extension in FL scenarios with low communication and computation costs, even without accessing the full model; (3) we adopt several accelerating and resource-efficient operators for fine-tuning LLMs with limited resources and the flexible pluggable sub-routines for interdisciplinary study. We conduct extensive experiments to validate the effectiveness of FS-LLM and benchmark advanced LLMs with state-of-the-art parameter-efficient fine-tuning algorithms in FL settings, which also yields valuable insights into federated fine-tuning LLMs for the research community. To facilitate further research and adoption, we release FS-LLM at https://github.com/alibaba/FederatedScope/tree/llm.
25. Implicit Style-Content Separation using B-LoRA
Authors: Yarden Frenkel, Yael Vinker, Ariel Shamir, Daniel Cohen-Or
Categories: cs.CV
Published: 2024-03-21
arXiv: 2403.14572v2
Abstract:
Image stylization involves manipulating the visual appearance and texture (style) of an image while preserving its underlying objects, structures, and concepts (content). The separation of style and content is essential for manipulating the image’s style independently from its content, ensuring a harmonious and visually pleasing result. Achieving this separation requires a deep understanding of both the visual and semantic characteristics of images, often necessitating the training of specialized models or employing heavy optimization. In this paper, we introduce B-LoRA, a method that leverages LoRA (Low-Rank Adaptation) to implicitly separate the style and content components of a single image, facilitating various image stylization tasks. By analyzing the architecture of SDXL combined with LoRA, we find that jointly learning the LoRA weights of two specific blocks (referred to as B-LoRAs) achieves style-content separation that cannot be achieved by training each B-LoRA independently. Consolidating the training into only two blocks and separating style and content allows for significantly improving style manipulation and overcoming overfitting issues often associated with model fine-tuning. Once trained, the two B-LoRAs can be used as independent components to allow various image stylization tasks, including image style transfer, text-based image stylization, consistent style generation, and style-content mixing.
26. Differentially Private Fine-tuning of Language Models
Authors: Da Yu, Saurabh Naik, Arturs Backurs, Sivakanth Gopi, Huseyin A. Inan, Gautam Kamath, Janardhan Kulkarni, Yin Tat Lee, Andre Manoel, Lukas Wutschitz, Sergey Yekhanin, Huishuai Zhang
Categories: cs.LG, cs.CL, cs.CR, stat.ML
Published: 2021-10-13
arXiv: 2110.06500v2
Abstract:
We give simpler, sparser, and faster algorithms for differentially private fine-tuning of large-scale pre-trained language models, which achieve the state-of-the-art privacy versus utility tradeoffs on many standard NLP tasks. We propose a meta-framework for this problem, inspired by the recent success of highly parameter-efficient methods for fine-tuning. Our experiments show that differentially private adaptations of these approaches outperform previous private algorithms in three important dimensions: utility, privacy, and the computational and memory cost of private training. On many commonly studied datasets, the utility of private models approaches that of non-private models. For example, on the MNLI dataset we achieve an accuracy of $87.8%$ using RoBERTa-Large and $83.5%$ using RoBERTa-Base with a privacy budget of $ε= 6.7$. In comparison, absent privacy constraints, RoBERTa-Large achieves an accuracy of $90.2%$. Our findings are similar for natural language generation tasks. Privately fine-tuning with DART, GPT-2-Small, GPT-2-Medium, GPT-2-Large, and GPT-2-XL achieve BLEU scores of 38.5, 42.0, 43.1, and 43.8 respectively (privacy budget of $ε= 6.8,δ=$ 1e-5) whereas the non-private baseline is $48.1$. All our experiments suggest that larger models are better suited for private fine-tuning: while they are well known to achieve superior accuracy non-privately, we find that they also better maintain their accuracy when privacy is introduced.
27. Fine-tuning with Very Large Dropout
Authors: Jianyu Zhang, Léon Bottou
Categories: cs.LG, cs.CV
Published: 2024-03-01
arXiv: 2403.00946v3
Abstract:
It is impossible today to pretend that the practice of machine learning is always compatible with the idea that training and testing data follow the same distribution. Several authors have recently used ensemble techniques to show how scenarios involving multiple data distributions are best served by representations that are both richer than those obtained by regularizing for the best in-distribution performance, and richer than those obtained under the influence of the implicit sparsity bias of common stochastic gradient procedures. This contribution investigates the use of very high dropout rates instead of ensembles to obtain such rich representations. Although training a deep network from scratch using such dropout rates is virtually impossible, fine-tuning a large pre-trained model under such conditions is not only possible but also achieves out-of-distribution performances that exceed those of both ensembles and weight averaging methods such as model soups. This result has practical significance because the importance of the fine-tuning scenario has considerably grown in recent years. This result also provides interesting insights on the nature of rich representations and on the intrinsically linear nature of fine-tuning a large network using a comparatively small dataset.
28. Legged Robots that Keep on Learning: Fine-Tuning Locomotion Policies in the Real World
Authors: Laura Smith, J. Chase Kew, Xue Bin Peng, Sehoon Ha, Jie Tan, Sergey Levine
Categories: cs.RO
Published: 2021-10-11
arXiv: 2110.05457v1
Abstract:
Legged robots are physically capable of traversing a wide range of challenging environments, but designing controllers that are sufficiently robust to handle this diversity has been a long-standing challenge in robotics. Reinforcement learning presents an appealing approach for automating the controller design process and has been able to produce remarkably robust controllers when trained in a suitable range of environments. However, it is difficult to predict all likely conditions the robot will encounter during deployment and enumerate them at training-time. What if instead of training controllers that are robust enough to handle any eventuality, we enable the robot to continually learn in any setting it finds itself in? This kind of real-world reinforcement learning poses a number of challenges, including efficiency, safety, and autonomy. To address these challenges, we propose a practical robot reinforcement learning system for fine-tuning locomotion policies in the real world. We demonstrate that a modest amount of real-world training can substantially improve performance during deployment, and this enables a real A1 quadrupedal robot to autonomously fine-tune multiple locomotion skills in a range of environments, including an outdoor lawn and a variety of indoor terrains.
29. Parameter-Efficient Fine-Tuning for Medical Text Summarization: A Comparative Study of Lora, Prompt Tuning, and Full Fine-Tuning
Authors: Ulugbek Shernazarov, Rostislav Svitsov, Bin Shi
Categories: cs.CL, cs.AI
Published: 2026-03-23
arXiv: 2603.21970v1
Abstract:
Fine-tuning large language models for domain-specific tasks such as medical text summarization demands substantial computational resources. Parameter-efficient fine-tuning (PEFT) methods offer promising alternatives by updating only a small fraction of parameters. This paper compares three adaptation approaches-Low-Rank Adaptation (LoRA), Prompt Tuning, and Full Fine-Tuning-across the Flan-T5 model family on the PubMed medical summarization dataset. Through experiments with multiple random seeds, we demonstrate that LoRA consistently outperforms full fine-tuning, achieving 43.52 +/- 0.18 ROUGE-1 on Flan-T5-Large with only 0.6% trainable parameters compared to 40.67 +/- 0.21 for full fine-tuning. Sensitivity analyses examine the impact of LoRA rank and prompt token count. Our findings suggest the low-rank constraint provides beneficial regularization, challenging assumptions about the necessity of full parameter updates. Code is available at https://github.com/eracoding/llm-medical-summarization
30. From inflation to a zero cosmological constant phase without fine tuning
Authors: E. I. Guendelman, A. B. Kaganovich
Categories: gr-qc
Published: 1997-09-24
arXiv: gr-qc/9709059v2
Abstract:
We show that it is possible to obtain inflation and also solve the cosmological constant problem. The theory is invariant under changes of the Lagrangian density $L$ to $L+const$. Then the constant part of a scalar field potential $V$ cannot be responsible for inflation. However, we show that inflation can be driven by a condensate of a four index field strength. A constraint appears which correlates this condensate to $V$. After a conformal transformation, the equations are the standard GR equations with an effective scalar field potential $V_{eff}$ which has generally an absolute minimum $V_{eff}=0$ independently of $V$ and without fine tuning. We also show that, after inflation, the usual reheating phase scenario (from oscillations around the absolute minimum) is possible.
31. Fine-Tuning a 7B Advisor on Free-Tier GPUs: An Adapter-Handoff Recipe and a Synthetic-Data Reliability Caution
Authors: Md Millat Hosen
Categories: cs.AI
Published: 2025-04-22
arXiv: 2504.15610v4
Abstract:
Fine-tuning a 7B language model for specialized advising is attractive in resource-constrained settings, but multi-epoch runs routinely exceed the wall-clock limits of the free-tier GPUs (Kaggle, Colab) such users rely on. We report two things. First, a practical recipe: a three-epoch QLoRA fine-tune of Mistral-7B-Instruct-v0.3 (4-bit NF4, LoRA rank 16, via Unsloth) completed across two free-tier 16 GB GPUs (Tesla P100 then T4) by checkpointing only the small LoRA adapter (41.9M parameters) and resuming on the second machine. Adapter-only handoff is sufficient – optimizer and scheduler state need not be transferred – so the binding constraint is per-step VRAM and per-session wall-clock, not aggregate compute. Second, and more importantly, an honest evaluation that returns a cautionary result. On a blind held-out comparison against the un-fine-tuned base model, the fine-tuned model scored higher on similarity to the synthetic training distribution (BERTScore F1 +0.063, a fidelity not quality signal) but lower on advising quality: a blind LLM-as-judge preferred the base model on 46% of prompts versus 18%, and a source-verified factuality audit found four confident errors from the fine-tuned model on policy-sensitive topics against zero for the base. Auditing the training data with the same method, we find this is not a fine-tuning artifact: each audited error is already present in the Gemini-generated training answers, and a random-sample audit finds verifiable errors in a sizable fraction of responses (28-40%; single-judge, n=40). The data is therefore sufficient to account for the errors, which we attribute to the synthetic-data pipeline rather than the adapter-handoff method. We release the dataset, adapter, cross-GPU notebooks, and full evaluation harness so every result reproduces on a single 16 GB GPU.
32. In-Context LoRA for Diffusion Transformers
Authors: Lianghua Huang, Wei Wang, Zhi-Fan Wu, Yupeng Shi, Huanzhang Dou, Chen Liang, Yutong Feng, Yu Liu, Jingren Zhou
Categories: cs.CV, cs.GR
Published: 2024-10-31
arXiv: 2410.23775v3
Abstract:
Recent research arXiv:2410.15027 has explored the use of diffusion transformers (DiTs) for task-agnostic image generation by simply concatenating attention tokens across images. However, despite substantial computational resources, the fidelity of the generated images remains suboptimal. In this study, we reevaluate and streamline this framework by hypothesizing that text-to-image DiTs inherently possess in-context generation capabilities, requiring only minimal tuning to activate them. Through diverse task experiments, we qualitatively demonstrate that existing text-to-image DiTs can effectively perform in-context generation without any tuning. Building on this insight, we propose a remarkably simple pipeline to leverage the in-context abilities of DiTs: (1) concatenate images instead of tokens, (2) perform joint captioning of multiple images, and (3) apply task-specific LoRA tuning using small datasets (e.g., 20~100 samples) instead of full-parameter tuning with large datasets. We name our models In-Context LoRA (IC-LoRA). This approach requires no modifications to the original DiT models, only changes to the training data. Remarkably, our pipeline generates high-fidelity image sets that better adhere to prompts. While task-specific in terms of tuning data, our framework remains task-agnostic in architecture and pipeline, offering a powerful tool for the community and providing valuable insights for further research on product-level task-agnostic generation systems. We release our code, data, and models at https://github.com/ali-vilab/In-Context-LoRA
33. LoftQ: LoRA-Fine-Tuning-Aware Quantization for Large Language Models
Authors: Yixiao Li, Yifan Yu, Chen Liang, Pengcheng He, Nikos Karampatziakis, Weizhu Chen, Tuo Zhao
Categories: cs.CL, cs.AI, cs.LG
Published: 2023-10-12
arXiv: 2310.08659v4
Abstract:
Quantization is an indispensable technique for serving Large Language Models (LLMs) and has recently found its way into LoRA fine-tuning. In this work we focus on the scenario where quantization and LoRA fine-tuning are applied together on a pre-trained model. In such cases it is common to observe a consistent gap in the performance on downstream tasks between full fine-tuning and quantization plus LoRA fine-tuning approach. In response, we propose LoftQ (LoRA-Fine-Tuning-aware Quantization), a novel quantization framework that simultaneously quantizes an LLM and finds a proper low-rank initialization for LoRA fine-tuning. Such an initialization alleviates the discrepancy between the quantized and full-precision model and significantly improves generalization in downstream tasks. We evaluate our method on natural language understanding, question answering, summarization, and natural language generation tasks. Experiments show that our method is highly effective and outperforms existing quantization methods, especially in the challenging 2-bit and 2/4-bit mixed precision regimes. The code is available on https://github.com/yxli2123/LoftQ.
34. LoRA vs Full Fine-tuning: An Illusion of Equivalence
Authors: Reece Shuttleworth, Jacob Andreas, Antonio Torralba, Pratyusha Sharma
Categories: cs.LG, cs.CL
Published: 2024-10-28
arXiv: 2410.21228v3
Abstract:
Fine-tuning is a crucial paradigm for adapting pre-trained large language models to downstream tasks. Recently, methods like Low-Rank Adaptation (LoRA) have been shown to effectively fine-tune LLMs with an extreme reduction in trainable parameters. But, \emph{are their learned solutions really equivalent?} We study how LoRA and full-finetuning change pre-trained models by analyzing the model’s weight matrices through the lens of their spectral properties. We find that LoRA and full fine-tuning yield weight matrices whose singular value decompositions exhibit very different structure: weight matrices trained with LoRA have new, high-ranking singular vectors, which we call \emph{intruder dimensions}, while those trained with full fine-tuning do not. Further, we extend the finding that LoRA forgets less than full fine-tuning and find its forgetting is vastly localized to the intruder dimension – by causally intervening on the intruder dimensions by changing their associated singular values post-fine-tuning, we show that they cause forgetting. Moreover, scaling them down significantly improves modeling of the pre-training distribution with a minimal drop in downstream task performance. Given this, we should expect accumulating intruder dimensions to be harmful and lead to more forgetting. This will be amplified during continual learning because of sequentially fine-tuning, and we show that LoRA models do accumulate intruder dimensions here tend to perform worse in this setting, emphasizing the practicality of our findings.
35. Learning to Select, Not Relearn: Hard-Routed Mixtures of Reasoning LoRAs
Authors: Seyed Alireza Molavi, Zhan Su, Yan Hu, Peyman Sheikholharam Mashhadi, Stefan Byttner, Prayag Tiwari
Categories: cs.AI, cs.LG
Published: 2026-06-30
arXiv: 2606.31413v1
Abstract:
Composing independently trained LoRA adapters into a single large language model is useful for multi-domain adaptation, especially when the original training data cannot be shared. A common approach is to use MoE-style routing over LoRA experts, but for frozen pretrained adapters, soft weighted combinations can change the unit-scale additive update under which each LoRA module was originally trained. We propose \textbf{Hard-Routed MoR-LoRA}, a two-stage framework for composing frozen reasoning LoRA experts through unit-scale hard selection. First, domain-specific LoRA adapters are trained independently using reinforcement learning from verifiable feedback to obtain reasoning experts. Then, all experts are frozen, reasoning traces are distilled from them, and only a lightweight shared router together with a small attention LoRA is trained for integration. The router selects exactly one expert per token using hard top-1 routing, while a straight-through estimator enables gradient-based training. Experiments across five benchmarks, multiple model scales, and additional model families show that Hard-Routed MoR-LoRA preserves expert behavior while requiring substantially fewer trainable parameters than soft-routing mixture baselines. Our analysis further shows that normalized soft mixtures often concentrate most routing mass on a single expert, suggesting that hard unit-scale routing provides a simple and efficient abstraction for frozen LoRA expert composition.
36. HDMoLE: Mixture of LoRA Experts with Hierarchical Routing and Dynamic Thresholds for Fine-Tuning LLM-based ASR Models
Authors: Bingshen Mu, Kun Wei, Qijie Shao, Yong Xu, Lei Xie
Categories: cs.SD, eess.AS
Published: 2024-09-30
arXiv: 2409.19878v3
Abstract:
Recent advancements in integrating Large Language Models (LLM) with automatic speech recognition (ASR) have performed remarkably in general domains. While supervised fine-tuning (SFT) of all model parameters is often employed to adapt pre-trained LLM-based ASR models to specific domains, it imposes high computational costs and notably reduces their performance in general domains. In this paper, we propose a novel parameter-efficient multi-domain fine-tuning method for adapting pre-trained LLM-based ASR models to multi-accent domains without catastrophic forgetting named \textit{HDMoLE}, which leverages hierarchical routing and dynamic thresholds based on combining low-rank adaptation (LoRA) with the mixer of experts (MoE) and can be generalized to any linear layer. Hierarchical routing establishes a clear correspondence between LoRA experts and accent domains, improving cross-domain collaboration among the LoRA experts. Unlike the static Top-K strategy for activating LoRA experts, dynamic thresholds can adaptively activate varying numbers of LoRA experts at each MoE layer. Experiments on the multi-accent and standard Mandarin datasets demonstrate the efficacy of HDMoLE. Applying HDMoLE to an LLM-based ASR model projector module achieves similar performance to full fine-tuning in the target multi-accent domains while using only 9.6% of the trainable parameters required for full fine-tuning and minimal degradation in the source general domain.
37. Cost-Aware Model Selection for Text Classification: Multi-Objective Trade-offs Between Fine-Tuned Encoders and LLM Prompting in Production
Authors: Alberto Andres Valdes Gonzalez
Categories: cs.CL
Published: 2026-02-06
arXiv: 2602.06370v1
Abstract:
Large language models (LLMs) such as GPT-4o and Claude Sonnet 4.5 have demonstrated strong capabilities in open-ended reasoning and generative language tasks, leading to their widespread adoption across a broad range of NLP applications. However, for structured text classification problems with fixed label spaces, model selection is often driven by predictive performance alone, overlooking operational constraints encountered in production systems. In this work, we present a systematic comparison of two contrasting paradigms for text classification: zero- and few-shot prompt-based large language models, and fully fine-tuned encoder-only architectures. We evaluate these approaches across four canonical benchmarks (IMDB, SST-2, AG News, and DBPedia), measuring predictive quality (macro F1), inference latency, and monetary cost. We frame model evaluation as a multi-objective decision problem and analyze trade-offs using Pareto frontier projections and a parameterized utility function reflecting different deployment regimes. Our results show that fine-tuned encoder-based models from the BERT family achieve competitive, and often superior, classification performance while operating at one to two orders of magnitude lower cost and latency compared to zero- and few-shot LLM prompting. Overall, our findings suggest that indiscriminate use of large language models for standard text classification workloads can lead to suboptimal system-level outcomes. Instead, fine-tuned encoders emerge as robust and efficient components for structured NLP pipelines, while LLMs are better positioned as complementary elements within hybrid architectures. We release all code, datasets, and evaluation protocols to support reproducibility and cost-aware NLP system design.
38. Partial Is Better Than All: Revisiting Fine-tuning Strategy for Few-shot Learning
Authors: Zhiqiang Shen, Zechun Liu, Jie Qin, Marios Savvides, Kwang-Ting Cheng
Categories: cs.CV, cs.AI, cs.LG
Published: 2021-02-08
arXiv: 2102.03983v1
Abstract:
The goal of few-shot learning is to learn a classifier that can recognize unseen classes from limited support data with labels. A common practice for this task is to train a model on the base set first and then transfer to novel classes through fine-tuning (Here fine-tuning procedure is defined as transferring knowledge from base to novel data, i.e. learning to transfer in few-shot scenario.) or meta-learning. However, as the base classes have no overlap to the novel set, simply transferring whole knowledge from base data is not an optimal solution since some knowledge in the base model may be biased or even harmful to the novel class. In this paper, we propose to transfer partial knowledge by freezing or fine-tuning particular layer(s) in the base model. Specifically, layers will be imposed different learning rates if they are chosen to be fine-tuned, to control the extent of preserved transferability. To determine which layers to be recast and what values of learning rates for them, we introduce an evolutionary search based method that is efficient to simultaneously locate the target layers and determine their individual learning rates. We conduct extensive experiments on CUB and mini-ImageNet to demonstrate the effectiveness of our proposed method. It achieves the state-of-the-art performance on both meta-learning and non-meta based frameworks. Furthermore, we extend our method to the conventional pre-training + fine-tuning paradigm and obtain consistent improvement.
39. A Rank Stabilization Scaling Factor for Fine-Tuning with LoRA
Authors: Damjan Kalajdzievski
Categories: cs.CL, cs.LG
Published: 2023-11-28
arXiv: 2312.03732v1
Abstract:
As large language models (LLMs) have become increasingly compute and memory intensive, parameter-efficient fine-tuning (PEFT) methods are now a common strategy to fine-tune LLMs. A popular PEFT method is Low-Rank Adapters (LoRA), which adds trainable low-rank “adapters” to selected layers. Each adapter consists of a low-rank matrix product, multiplicatively scaled by a rank-dependent factor. This scaling factor, which divides adapters by a factor of the rank, results in slowed learning and stunted performance for LoRA with higher-rank adapters. Consequently, the use of LoRA in practice has generally been limited to very low ranks. In this work, we study the impact of the scaling factor on the learning process and prove that LoRA adapters should be divided by a factor of the square root of the rank. Modifying LoRA with the appropriate scaling factor, which we call the rank-stabilized LoRA (rsLoRA) method, easily provides for a fine-tuning compute/performance trade-off, where larger ranks can be used to trade off increased computational resources during training for better fine-tuning performance, with no change in inference computing cost.
40. Put the Space of LoRA Initialization to the Extreme to Preserve Pre-trained Knowledge
Authors: Pengwei Tang, Xiaolin Hu, Yong Liu, Lizhong Ding, Dongjie Zhang, Xing Wu, Debing Zhang
Categories: cs.CL
Published: 2025-03-04
arXiv: 2503.02659v2
Abstract:
Low-Rank Adaptation (LoRA) is the leading parameter-efficient fine-tuning method for Large Language Models (LLMs), but it still suffers from catastrophic forgetting. Recent work has shown that specialized LoRA initialization can alleviate catastrophic forgetting. There are currently two approaches to LoRA initialization aimed at preventing knowledge forgetting during fine-tuning: (1) making residual weights close to pre-trained weights, and (2) ensuring the space of LoRA initialization is orthogonal to pre-trained knowledge. The former is what current methods strive to achieve, while the importance of the latter is not sufficiently recognized. We find that the space of LoRA initialization is the key to preserving pre-trained knowledge rather than the residual weights. Existing methods like MiLoRA propose making the LoRA initialization space orthogonal to pre-trained weights. However, MiLoRA utilizes the null space of pre-trained weights. Compared to pre-trained weights, the input activations of pre-trained knowledge take into account the parameters of all previous layers as well as the input data, while pre-trained weights only contain information from the current layer. Moreover, we find that the effective ranks of input activations are much smaller than those of pre-trained weights. Thus, the null space of activations is more accurate and contains less pre-trained knowledge information compared to that of weights. Based on these, we introduce LoRA-Null, our proposed method that initializes LoRA in the null space of activations. Experimental results show that LoRA-Null effectively preserves the pre-trained world knowledge of LLMs while achieving good fine-tuning performance, as evidenced by extensive experiments. Code is available at {https://github.com/HungerPWAY/LoRA-Null}.
41. SC-LoRA: Balancing Efficient Fine-tuning and Knowledge Preservation via Subspace-Constrained LoRA
Authors: Minrui Luo, Fuhang Kuang, Yu Wang, Zirui Liu, Tianxing He
Categories: cs.LG, cs.AI
Published: 2025-05-29
arXiv: 2505.23724v3
Abstract:
Parameter-Efficient Fine-Tuning (PEFT) methods, particularly Low-Rank Adaptation (LoRA), are indispensable for efficiently customizing Large Language Models (LLMs). However, vanilla LoRA suffers from slow convergence speed and knowledge forgetting problems. Recent studies have leveraged the power of designed LoRA initialization, to enhance the fine-tuning efficiency, or to preserve knowledge in the pre-trained LLM. However, none of these works can address the two cases at the same time. To this end, we introduce Subspace-Constrained LoRA (SC-LoRA), a novel LoRA initialization framework engineered to navigate the trade-off between efficient fine-tuning and knowledge preservation. We achieve this by constraining the output of trainable LoRA adapters in a low-rank subspace, where the context information of fine-tuning data is most preserved while the context information of preserved knowledge is least retained, in a balanced way. Such constraint enables the trainable weights to primarily focus on the main features of fine-tuning data while avoiding damaging the preserved knowledge features. We provide theoretical analysis on our method, and conduct extensive experiments including safety preservation and world knowledge preservation, on various downstream tasks. In our experiments, SC-LoRA succeeds in delivering superior fine-tuning performance while markedly diminishing knowledge forgetting, surpassing contemporary LoRA initialization methods.
42. FinLoRA: Benchmarking LoRA Methods for Fine-Tuning LLMs on Financial Datasets
Authors: Dannong Wang, Jaisal Patel, Daochen Zha, Steve Y. Yang, Xiao-Yang Liu
Categories: cs.CE, cs.AI
Published: 2025-05-26
arXiv: 2505.19819v1
Abstract:
Low-rank adaptation (LoRA) methods show great potential for scaling pre-trained general-purpose Large Language Models (LLMs) to hundreds or thousands of use scenarios. However, their efficacy in high-stakes domains like finance is rarely explored, e.g., passing CFA exams and analyzing SEC filings. In this paper, we present the open-source FinLoRA project that benchmarks LoRA methods on both general and highly professional financial tasks. First, we curated 19 datasets covering diverse financial applications; in particular, we created four novel XBRL analysis datasets based on 150 SEC filings. Second, we evaluated five LoRA methods and five base LLMs. Finally, we provide extensive experimental results in terms of accuracy, F1, and BERTScore and report computational cost in terms of time and GPU memory during fine-tuning and inference stages. We find that LoRA methods achieved substantial performance gains of 36% on average over base models. Our FinLoRA project provides an affordable and scalable approach to democratize financial intelligence to the general public. Datasets, LoRA adapters, code, and documentation are available at https://github.com/Open-Finance-Lab/FinLoRA
43. PLoP: Precise LoRA Placement for Efficient Finetuning of Large Models
Authors: Soufiane Hayou, Nikhil Ghosh, Bin Yu
Categories: cs.LG, cs.CL, stat.ML
Published: 2025-06-25
arXiv: 2506.20629v1
Abstract:
Low-Rank Adaptation (LoRA) is a widely used finetuning method for large models. Its small memory footprint allows practitioners to adapt large models to specific tasks at a fraction of the cost of full finetuning. Different modifications have been proposed to enhance its efficiency by, for example, setting the learning rate, the rank, and the initialization. Another improvement axis is adapter placement strategy: when using LoRA, practitioners usually pick module types to adapt with LoRA, such as Query and Key modules. Few works have studied the problem of adapter placement, with nonconclusive results: original LoRA paper suggested placing adapters in attention modules, while other works suggested placing them in the MLP modules. Through an intuitive theoretical analysis, we introduce PLoP (Precise LoRA Placement), a lightweight method that allows automatic identification of module types where LoRA adapters should be placed, given a pretrained model and a finetuning task. We demonstrate that PLoP consistently outperforms, and in the worst case competes, with commonly used placement strategies through comprehensive experiments on supervised finetuning and reinforcement learning for reasoning.
44. LUNE: Efficient LLM Unlearning via LoRA Fine-Tuning with Negative Examples
Authors: Yezi Liu, Hanning Chen, Wenjun Huang, Yang Ni, Mohsen Imani
Categories: cs.LG, cs.CL
Published: 2025-12-08
arXiv: 2512.07375v1
Abstract:
Large language models (LLMs) possess vast knowledge acquired from extensive training corpora, but they often cannot remove specific pieces of information when needed, which makes it hard to handle privacy, bias mitigation, and knowledge correction. Traditional model unlearning approaches require computationally expensive fine-tuning or direct weight editing, making them impractical for real-world deployment. In this work, we introduce LoRA-based Unlearning with Negative Examples (LUNE), a lightweight framework that performs negative-only unlearning by updating only low-rank adapters while freezing the backbone, thereby localizing edits and avoiding disruptive global changes. Leveraging Low-Rank Adaptation (LoRA), LUNE targets intermediate representations to suppress (or replace) requested knowledge with an order-of-magnitude lower compute and memory than full fine-tuning or direct weight editing. Extensive experiments on multiple factual unlearning tasks show that LUNE: (I) achieves effectiveness comparable to full fine-tuning and memory-editing methods, and (II) reduces computational cost by about an order of magnitude.
45. Adaptive LoRA Experts Allocation and Selection for Federated Fine-Tuning
Authors: Lei Wang, Jieming Bian, Letian Zhang, Jie Xu
Categories: cs.LG
Published: 2025-09-18
arXiv: 2509.15087v2
Abstract:
Large Language Models (LLMs) have demonstrated impressive capabilities across various tasks, but fine-tuning them for domain-specific applications often requires substantial domain-specific data that may be distributed across multiple organizations. Federated Learning (FL) offers a privacy-preserving solution, but faces challenges with computational constraints when applied to LLMs. Low-Rank Adaptation (LoRA) has emerged as a parameter-efficient fine-tuning approach, though a single LoRA module often struggles with heterogeneous data across diverse domains. This paper addresses two critical challenges in federated LoRA fine-tuning: 1. determining the optimal number and allocation of LoRA experts across heterogeneous clients, and 2. enabling clients to selectively utilize these experts based on their specific data characteristics. We propose FedLEASE (Federated adaptive LoRA Expert Allocation and SElection), a novel framework that adaptively clusters clients based on representation similarity to allocate and train domain-specific LoRA experts. It also introduces an adaptive top-$M$ Mixture-of-Experts mechanism that allows each client to select the optimal number of utilized experts. Our extensive experiments on diverse benchmark datasets demonstrate that FedLEASE significantly outperforms existing federated fine-tuning approaches in heterogeneous client settings while maintaining communication efficiency.
46. KD-LoRA: A Hybrid Approach to Efficient Fine-Tuning with LoRA and Knowledge Distillation
Authors: Rambod Azimi, Rishav Rishav, Marek Teichmann, Samira Ebrahimi Kahou
Categories: cs.CL, cs.AI, cs.LG
Published: 2024-10-28
arXiv: 2410.20777v1
Abstract:
Large language models (LLMs) have demonstrated remarkable performance across various downstream tasks. However, the high computational and memory requirements of LLMs are a major bottleneck. To address this, parameter-efficient fine-tuning (PEFT) methods such as low-rank adaptation (LoRA) have been proposed to reduce computational costs while ensuring minimal loss in performance. Additionally, knowledge distillation (KD) has been a popular choice for obtaining compact student models from teacher models. In this work, we present KD-LoRA, a novel fine-tuning method that combines LoRA with KD. Our results demonstrate that KD-LoRA achieves performance comparable to full fine-tuning (FFT) and LoRA while significantly reducing resource requirements. Specifically, KD-LoRA retains 98% of LoRA’s performance on the GLUE benchmark, while being 40% more compact. Additionally, KD-LoRA reduces GPU memory usage by 30% compared to LoRA, while decreasing inference time by 30% compared to both FFT and LoRA. We evaluate KD-LoRA across three encoder-only models: BERT, RoBERTa, and DeBERTaV3. Code is available at https://github.com/rambodazimi/KD-LoRA.
47. CE-LoRA: Computation-Efficient LoRA Fine-Tuning for Language Models
Authors: Guanduo Chen, Yutong He, Yipeng Hu, Kun Yuan, Binhang Yuan
Categories: cs.LG
Published: 2025-02-03
arXiv: 2502.01378v1
Abstract:
Large Language Models (LLMs) demonstrate exceptional performance across various tasks but demand substantial computational resources even for fine-tuning computation. Although Low-Rank Adaptation (LoRA) significantly alleviates memory consumption during fine-tuning, its impact on computational cost reduction is limited. This paper identifies the computation of activation gradients as the primary bottleneck in LoRA’s backward propagation and introduces the Computation-Efficient LoRA (CE-LoRA) algorithm, which enhances computational efficiency while preserving memory efficiency. CE-LoRA leverages two key techniques: Approximated Matrix Multiplication, which replaces dense multiplications of large and complete matrices with sparse multiplications involving only critical rows and columns, and the Double-LoRA technique, which reduces error propagation in activation gradients. Theoretically, CE-LoRA converges at the same rate as LoRA, $ \mathcal{O}(1/\sqrt{T}) $, where $T$ is the number of iteartions. Empirical evaluations confirm that CE-LoRA significantly reduces computational costs compared to LoRA without notable performance degradation.
48. Analyzing Quality-Latency-Resource Trade-offs in a Technical Documentation RAG Assistant Using LoRA Adaptation
Authors: Evgenii Palnikov, Elizaveta Gavrilova
Categories: cs.CL, cs.IR, cs.LG
Published: 2026-05-27
arXiv: 2605.28222v1
Abstract:
We study quality-latency-resource trade-offs in a documentation-grounded retrieval-augmented generation (RAG) system that uses Low-Rank Adaptation (LoRA) of the generator. We build a manually verified benchmark of 5,144 question-answer pairs over the official Kubernetes documentation and combine it with a fixed hybrid-retrieval pipeline (BGE-M3 dense, BGE-M3 native sparse, Reciprocal Rank Fusion, cross-encoder reranking). Over this benchmark we ablate 20 LoRA configurations on Llama-3.2-3B-Instruct and Llama-3.1-8B-Instruct across rank and target-module choices, and evaluate each on token-level F1, LLM-judged groundedness and correctness (pass@4), inference latency, inference memory, and training cost, all reported with bootstrap 95% confidence intervals. Pareto analysis shows that LoRA adapters acting only on the q and v attention projections consistently dominate the front, while the 3B/8B choice mainly defines operating regime. A param-matched control comparison further indicates that the q/v advantage is structural rather than purely parametric. The benchmark, selected adapters, and code are available at https://github.com/EugPal/rag-lora-tradeoffs.
49. LoRA Fine-Tuning Without GPUs: A CPU-Efficient Meta-Generation Framework for LLMs
Authors: Reza Arabpour, Haitz Sáez de Ocáriz Borde, Anastasis Kratsios
Categories: cs.LG, cs.AI, cs.CL, stat.ML
Published: 2025-07-02
arXiv: 2507.01806v1
Abstract:
Low-Rank Adapters (LoRAs) have transformed the fine-tuning of Large Language Models (LLMs) by enabling parameter-efficient updates. However, their widespread adoption remains limited by the reliance on GPU-based training. In this work, we propose a theoretically grounded approach to LoRA fine-tuning designed specifically for users with limited computational resources, particularly those restricted to standard laptop CPUs. Our method learns a meta-operator that maps any input dataset, represented as a probability distribution, to a set of LoRA weights by leveraging a large bank of pre-trained adapters for the Mistral-7B-Instruct-v0.2 model. Instead of performing new gradient-based updates, our pipeline constructs adapters via lightweight combinations of existing LoRAs directly on CPU. While the resulting adapters do not match the performance of GPU-trained counterparts, they consistently outperform the base Mistral model on downstream tasks, offering a practical and accessible alternative to traditional GPU-based fine-tuning.
50. LoRA vs. Full Fine-Tuning: A Theoretical Perspective
Authors: Ali Zindari, Rotem Mulayoff, Sebastian U. Stich
Categories: cs.LG
Published: 2026-05-18
arXiv: 2605.19018v1
Abstract:
Fine-tuning adapts a pre-trained model to downstream tasks using a small amount of labeled data. Low-Rank Adaptation (LoRA) is an efficient fine-tuning method that reduces memory and computation costs while often achieving performance close to full fine-tuning. Despite its widespread use, the theoretical behavior of LoRA is not yet well understood. In this paper, we study LoRA in a simple linear regression setting and compare its excess risk with that of full fine-tuning. Our analysis identifies regimes in which LoRA achieves lower excess risk than full fine-tuning in both overdetermined and underdetermined settings. Specifically, our theory predicts that LoRA can outperform full fine-tuning when the difference between the pretraining and the downstream tasks is effectively low-rank. We further show how the choice of LoRA rank affects generalization performance, explaining why using a very small rank can improve test accuracy in certain settings, even though it limits model expressivity. Finally, we support our theoretical results with experiments on practical tasks, suggesting that the identified tradeoffs and insights extend beyond linear regression.