NSTL国家科技图书文献中心

1. ERAT-DLoRA: Parameter-efficient tuning with enhanced range adaptation in time and depth aware dynamic LoRA NSTL国家科技图书文献中心

Luo D. | Zheng K.... - 《Neurocomputing》 - 2025,614(Jan.21) - 1.1～1.12 - 共12页

摘要： by traditional fine-tuning* procedures that are both* | © 2024Despite their potential, the industrial | deployment of large language models (LLMs) is constrained | resource-intensive and time-consuming. Low-Rank | Adaptation (LoRA) has emerged as a pioneering methodology

关键词： Fine-tuning | LoRA | Parameter-efficient

2. EC-PEFT: An Expertise-Centric Parameter-Efficient Fine-Tuning Framework for Large Language Models NSTL国家科技图书文献中心

Yimeng Zhang | Xuelin Cheng ... - 《PRICAI 2024,Part II》 - Pacific Rim International Conference on Artificial Intelligence - 2025, - 296～307 - 共12页

摘要：, parameter-efficient* fine-tuning (PEFT) methods have gained* | Expertise-Centric Parameter-Efficient* Fine-Tuning (EC-PEFT* | parameter-efficient tuning modules as experts. By | -efficient* tuning modules: EC-LoRA and EC-Adapter. We* | increasing attention. However, fine-tuning* methods often*

关键词： Parameter-Efficient fine-Tuning | Mixture-of-Experts | Contrastive learning

3. Leveraging Parameter-Efficient Fine-Tuning for Multilingual Abstractive Summarization NSTL国家科技图书文献中心

Jialun Shen | Yusong Wang - 《Natural Language Processing and Chinese Computing,Part III》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 293～303 - 共11页

摘要： and benefits of parameter-efficient fine-tuning | encourage the adoption of parameter-efficient fine-tuning | In recent years, parameter-efficient* fine* | performance of two representative parameter-efficient* fine* | -tuning* methods have gained attention as an alternative*

关键词： Transfer learning | Parameter-efficient fine-tuning | Multilingual abstractive summarization | Pre-trained language models

4. Reparameterization-Based Parameter-Efficient Fine-Tuning Methods for Large Language Models: A Systematic Survey NSTL国家科技图书文献中心

Zezhou Chen | Zhaoxiang Liu ... - 《Natural Language Processing and Chinese Computing,Part III》 - CCF International Conference on Natural Language Processing and Chinese Computing - 2025, - 107～118 - 共12页

摘要：-Efficient* Fine-Tuning (PEFT) methods, especially* | exploit the potential of LLMs, fine-tuning* LLMs on* | , traditional full fine-tuning* methods pose significant* | computational challenges, prompting the emergence of Parameter | parameter complexity, GPU memory consumption, training

关键词： Large language models | Parameter-Efficient fine-Tuning | Reparameterization

5. Introducing Routing Functions to Vision-Language Parameter-Efficient Fine-Tuning with Low-Rank Bottlenecks NSTL国家科技图书文献中心

Tingyu Qu | Tinne Tuytelaars ... - 《Computer Vision - ECCV 2024,Part LXXXVIII》 - European Conference on Computer Vision - 2025, - 291～308 - 共18页

摘要：Mainstream parameter-efficient fine-tuning | fine-tuning* a pre-trained multimodal model such as* | (PEFT) methods, such as LoRA or Adapter, project a | model's hidden states to a lower dimension, allowing pre | -trained models to adapt to new data through this low

关键词： Vision and language | Parameter-Efficient fine-Tuning | Low-Rank approximation

6. Deconfounded Causality-Aware Parameter-Efficient Fine-Tuning for Problem-Solving Improvement of LLMs NSTL国家科技图书文献中心

Ruoyu Wang | Xiaoxuan Li ... - 《Web Information Systems Engineering - WISE 2024,Part IV》 - International Conference on Web Information Systems Engineering - 2025, - 161～176 - 共16页

摘要： novel parameter-efficient* fine-tuning (PEFT) method to* | results to other fine-tuning* methods. This demonstrates* | Large Language Models (LLMs) have demonstrated | remarkable efficiency in tackling various tasks based on | human instructions, but studies reveal that they often

关键词： Parameter-Efficient fine-Tuning (PEFT) | Causality | Large language models

7. VL-MPFT: Multitask Parameter-Efficient Fine-Tuning for Visual-Language Pre-trained Models via Task-Adaptive Masking NSTL国家科技图书文献中心

Min Zhu | Guanming Liu ... - 《Pattern Recognition and Computer Vision,Part V》 - Chinese Conference on Pattern Recognition and Computer Vision - 2025, - 379～394 - 共16页

摘要：Parameter-efficient fine-tuning (PEFT) has | , existing parameter-efficient* fine-tuning methods for VLMs* | parameter-efficient fine-tuning for visual-language pre | selectively updating a small subset for fine-tuning. However | fine-tuning* accordingly. In this paper, the proposed*

关键词： Multimodel learning | Multitask learning | Parameter-efficient finetune-tuning

8. A Federated Learning-Friendly Approach for Parameter-Efficient Fine-Tuning of SAM in 3D Segmentation NSTL国家科技图书文献中心

Mothilal Asokan | Joseph Geo Benjamin ... - 《Medical Image Computing and Computer Assisted Intervention - MICCAI 2024 Workshops》 - International Conference on Medical Image Computing and Computer Assisted Intervention | International Workshop on Interpretability of Machine Intelligence in Medical Image Computing | Workshop on Distributed,Collaborative,and Federated Learning | Embodied AI and Robotics for HealTHcare | International Skin Imaging Collaboration - 2025, - 226～235 - 共10页

摘要： combining the strengths of Parameter-Efficient Fine-tuning | . In this work, we address this problem of 'efficient | specific layers to be federated that are very efficient | adaptation is beneficial because fine-tuning* on small* | to full fine-tuning* while increasing performance*

关键词： Federated learning | Foundation model | 3D medical image segmentation | Parameter-Efficient fine-Tuning

9. Probing the Efficacy of Federated Parameter-Efficient Fine-Tuning of Vision Transformers for Medical Image Classification NSTL国家科技图书文献中心

Naif Alkhunaizi | Faris Almalik ... - 《Medical Image Computing and Computer Assisted Intervention - MICCAI 2024 Workshops》 - International Conference on Medical Image Computing and Computer Assisted Intervention | International Workshop on Interpretability of Machine Intelligence in Medical Image Computing | Workshop on Distributed,Collaborative,and Federated Learning | Embodied AI and Robotics for HealTHcare | International Skin Imaging Collaboration - 2025, - 236～245 - 共10页

摘要：-efficient* fine-tuning (PEFT) to reduce the communication* | transformer models, fine-tuning* these models for various* | privacy constraints exacerbate this fine-tuning* problem* | -tuning* of pre-trained models. Moreover, the large size* | of these models necessitates the use of parameter

关键词： Vision transformers | Parameter-Efficient fine-tuning | Out-of-Domain transfer | Federated learning

10. Enhancing text understanding of decoder-based model by leveraging parameter-efficient fine-tuning method NSTL国家科技图书文献中心

Wasif,Feroze | Shaohuan,Cheng ... - 《Neural computing & applications》 - 2025,37(9) - 6899～6913 - 共15页

摘要： propose a parameter-efficient* fine-tuning framework that* | Machine reading comprehension (MRC) is a | fundamental natural language understanding task in natural | language processing, which aims to comprehend the text of | a given passage and answer questions based on it

摘要： by traditional fine-tuning* procedures that are both* | © 2024Despite their potential, the industrial | deployment of large language models (LLMs) is constrained | resource-intensive and time-consuming. Low-Rank | Adaptation (LoRA) has emerged as a pioneering methodology

关键词： Fine-tuning | LoRA | Parameter-efficient

关键词： Parameter-Efficient fine-Tuning | Mixture-of-Experts | Contrastive learning

摘要： and benefits of parameter-efficient fine-tuning | encourage the adoption of parameter-efficient fine-tuning | In recent years, parameter-efficient* fine* | performance of two representative parameter-efficient* fine* | -tuning* methods have gained attention as an alternative*

关键词： Transfer learning | Parameter-efficient fine-tuning | Multilingual abstractive summarization | Pre-trained language models

摘要：-Efficient* Fine-Tuning (PEFT) methods, especially* | exploit the potential of LLMs, fine-tuning* LLMs on* | , traditional full fine-tuning* methods pose significant* | computational challenges, prompting the emergence of Parameter | parameter complexity, GPU memory consumption, training

关键词： Large language models | Parameter-Efficient fine-Tuning | Reparameterization

摘要：Mainstream parameter-efficient fine-tuning | fine-tuning* a pre-trained multimodal model such as* | (PEFT) methods, such as LoRA or Adapter, project a | model's hidden states to a lower dimension, allowing pre | -trained models to adapt to new data through this low

关键词： Vision and language | Parameter-Efficient fine-Tuning | Low-Rank approximation

摘要： novel parameter-efficient* fine-tuning (PEFT) method to* | results to other fine-tuning* methods. This demonstrates* | Large Language Models (LLMs) have demonstrated | remarkable efficiency in tackling various tasks based on | human instructions, but studies reveal that they often

关键词： Parameter-Efficient fine-Tuning (PEFT) | Causality | Large language models

摘要：Parameter-efficient fine-tuning (PEFT) has | , existing parameter-efficient* fine-tuning methods for VLMs* | parameter-efficient fine-tuning for visual-language pre | selectively updating a small subset for fine-tuning. However | fine-tuning* accordingly. In this paper, the proposed*

关键词： Multimodel learning | Multitask learning | Parameter-efficient finetune-tuning

摘要： combining the strengths of Parameter-Efficient Fine-tuning | . In this work, we address this problem of 'efficient | specific layers to be federated that are very efficient | adaptation is beneficial because fine-tuning* on small* | to full fine-tuning* while increasing performance*

关键词： Federated learning | Foundation model | 3D medical image segmentation | Parameter-Efficient fine-Tuning

关键词： Vision transformers | Parameter-Efficient fine-tuning | Out-of-Domain transfer | Federated learning

摘要： propose a parameter-efficient* fine-tuning framework that* | Machine reading comprehension (MRC) is a | fundamental natural language understanding task in natural | language processing, which aims to comprehend the text of | a given passage and answer questions based on it

关键词： Natural language understanding | MRC question answering | Parameter-efficient fine-tuning | Large language models

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： by traditional fine-tuning procedures that are both | © 2024Despite their potential, the industrial | deployment of large language models (LLMs) is constrained | resource-intensive and time-consuming. Low-Rank | Adaptation (LoRA) has emerged as a pioneering methodology

关键词： Fine-tuning | LoRA | Parameter-efficient

摘要：, parameter-efficient fine-tuning (PEFT) methods have gained | Expertise-Centric Parameter-Efficient Fine-Tuning (EC-PEFT | parameter-efficient tuning modules as experts. By | -efficient tuning modules: EC-LoRA and EC-Adapter. We | increasing attention. However, fine-tuning methods often

关键词： Parameter-Efficient fine-Tuning | Mixture-of-Experts | Contrastive learning

摘要： and benefits of parameter-efficient fine-tuning | encourage the adoption of parameter-efficient fine-tuning | In recent years, parameter-efficient fine | performance of two representative parameter-efficient fine | -tuning methods have gained attention as an alternative

关键词： Transfer learning | Parameter-efficient fine-tuning | Multilingual abstractive summarization | Pre-trained language models

摘要：-Efficient Fine-Tuning (PEFT) methods, especially | exploit the potential of LLMs, fine-tuning LLMs on | , traditional full fine-tuning methods pose significant | computational challenges, prompting the emergence of Parameter | parameter complexity, GPU memory consumption, training

关键词： Large language models | Parameter-Efficient fine-Tuning | Reparameterization

摘要：Mainstream parameter-efficient fine-tuning | fine-tuning a pre-trained multimodal model such as | (PEFT) methods, such as LoRA or Adapter, project a | model's hidden states to a lower dimension, allowing pre | -trained models to adapt to new data through this low

关键词： Vision and language | Parameter-Efficient fine-Tuning | Low-Rank approximation

摘要： novel parameter-efficient fine-tuning (PEFT) method to | results to other fine-tuning methods. This demonstrates | Large Language Models (LLMs) have demonstrated | remarkable efficiency in tackling various tasks based on | human instructions, but studies reveal that they often

关键词： Parameter-Efficient fine-Tuning (PEFT) | Causality | Large language models

摘要：Parameter-efficient fine-tuning (PEFT) has | , existing parameter-efficient fine-tuning methods for VLMs | parameter-efficient fine-tuning for visual-language pre | selectively updating a small subset for fine-tuning. However | fine-tuning accordingly. In this paper, the proposed

关键词： Multimodel learning | Multitask learning | Parameter-efficient finetune-tuning

摘要： combining the strengths of Parameter-Efficient Fine-tuning | . In this work, we address this problem of 'efficient | specific layers to be federated that are very efficient | adaptation is beneficial because fine-tuning on small | to full fine-tuning while increasing performance

关键词： Federated learning | Foundation model | 3D medical image segmentation | Parameter-Efficient fine-Tuning

摘要：-efficient fine-tuning (PEFT) to reduce the communication | transformer models, fine-tuning these models for various | privacy constraints exacerbate this fine-tuning problem | -tuning of pre-trained models. Moreover, the large size | of these models necessitates the use of parameter

关键词： Vision transformers | Parameter-Efficient fine-tuning | Out-of-Domain transfer | Federated learning

摘要： propose a parameter-efficient fine-tuning framework that | Machine reading comprehension (MRC) is a | fundamental natural language understanding task in natural | language processing, which aims to comprehend the text of | a given passage and answer questions based on it

关键词： Natural language understanding | MRC question answering | Parameter-efficient fine-tuning | Large language models

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： by traditional fine-tuning* procedures that are both* | © 2024Despite their potential, the industrial | deployment of large language models (LLMs) is constrained | resource-intensive and time-consuming. Low-Rank | Adaptation (LoRA) has emerged as a pioneering methodology

摘要： and benefits of parameter-efficient fine-tuning | encourage the adoption of parameter-efficient fine-tuning | In recent years, parameter-efficient* fine* | performance of two representative parameter-efficient* fine* | -tuning* methods have gained attention as an alternative*

摘要：-Efficient* Fine-Tuning (PEFT) methods, especially* | exploit the potential of LLMs, fine-tuning* LLMs on* | , traditional full fine-tuning* methods pose significant* | computational challenges, prompting the emergence of Parameter | parameter complexity, GPU memory consumption, training

摘要：Mainstream parameter-efficient fine-tuning | fine-tuning* a pre-trained multimodal model such as* | (PEFT) methods, such as LoRA or Adapter, project a | model's hidden states to a lower dimension, allowing pre | -trained models to adapt to new data through this low

摘要： novel parameter-efficient* fine-tuning (PEFT) method to* | results to other fine-tuning* methods. This demonstrates* | Large Language Models (LLMs) have demonstrated | remarkable efficiency in tackling various tasks based on | human instructions, but studies reveal that they often

摘要：Parameter-efficient fine-tuning (PEFT) has | , existing parameter-efficient* fine-tuning methods for VLMs* | parameter-efficient fine-tuning for visual-language pre | selectively updating a small subset for fine-tuning. However | fine-tuning* accordingly. In this paper, the proposed*

摘要： combining the strengths of Parameter-Efficient Fine-tuning | . In this work, we address this problem of 'efficient | specific layers to be federated that are very efficient | adaptation is beneficial because fine-tuning* on small* | to full fine-tuning* while increasing performance*

摘要： propose a parameter-efficient* fine-tuning framework that* | Machine reading comprehension (MRC) is a | fundamental natural language understanding task in natural | language processing, which aims to comprehend the text of | a given passage and answer questions based on it