Results 61 to 70 of about 124,133 (290)

GeoLoRA: Geometric integration for parameter efficient fine-tuning

CoRR
Low-Rank Adaptation (LoRA) has become a widely used method for parameter-efficient fine-tuning of large-scale, pre-trained neural networks. However, LoRA and its extensions face several challenges, including the need for rank adaptivity, robustness, and computational efficiency during the fine-tuning process. We introduce GeoLoRA, a novel approach that
Steffen Schotthöfer, Emanuele Zangrando, Gianluca Ceruti, Francesco Tudisco, Jonas Kusch   +4 more
openaire   +4 more sources

Screening Routine Clinical Notes for Epilepsy Surgery Candidates Using Large Language Models

Annals of Clinical and Translational Neurology, EarlyView.
ABSTRACT Objective Epilepsy surgery is severely underutilized despite proven efficacy, with substantial under‐referral of eligible patients in routine clinical practice. This study evaluated the potential role of large language models (LLMs) as decision‐support tools for screening unstructured clinical notes to identify epilepsy surgery candidates and ...
Uriel Fennig, Nadav Amir, Maya Schiller, Roni Loebenstein, Johnatan Nissan, Marina Boxer, Shany Guly Gofrit, Gal Goshen, Sándor Beniczky, Nicola Maggio   +9 more
wiley   +1 more source

Know Where You're Going: Meta-Learning for Parameter-Efficient Fine-Tuning

, 2022
A recent family of techniques, dubbed lightweight fine-tuning methods, facilitates parameter-efficient transfer learning by updating only a small set of additional parameters while keeping the parameters of the pretrained language model frozen.
Gheini, Mozhdeh, May, Jonathan, Ma, Xuezhe   +2 more
core  

Parameter-Efficient Fine-Tuning with Circulant and Diagonal Vectors

Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence
Foundation models have achieved tremendous success in different domains. However, their huge computation and storage complexity make these models difficult to fine-tune and also less applicable in practice. Recent study shows training in Fourier domain can be an effective fine-tuning method in terms of both model performance and number of training ...
Xinyu Ding, Lexuan Chen, Siyu Liao, Zhongfeng Wang   +3 more
openaire   +2 more sources

Experimental Evaluation of 100Cr6 Steel Microindented Surfaces Under Lubricated Nonconformal Point Contacts

Advanced Engineering Materials, EarlyView.
The tribological behavior of 100Cr6 steel spheres textured via Vickers microindentation is evaluated under lubricated sliding by varying both dimple size and density. Fine and dense textures significantly reduce friction across all lubrication regimes, while large dimples increase it.
Farideh Davoodi, Pasquale Guglielmi, Gianfranco Palumbo, Giuseppe Carbone   +3 more
wiley   +1 more source

Parameter-Efficient Continual Fine-Tuning: A Survey

The emergence of large pre-trained networks has revolutionized the AI field, unlocking new possibilities and achieving unprecedented performance. However, these models inherit a fundamental limitation from traditional Machine Learning approaches: their strong dependence on the \textit{i.i.d.} assumption hinders their adaptability to dynamic learning ...
Eric Nuertey Coleman, Luigi Quarantiello, Ziyue Liu, Qinwen Yang, Samrat Mukherjee, Julio Hurtado, Vincenzo Lomonaco   +6 more
openaire   +2 more sources

What Do Large Language Models Know About Materials?

Advanced Engineering Materials, EarlyView.
If large language models (LLMs) are to be used inside the material discovery and engineering process, they must be benchmarked for the accurateness of intrinsic material knowledge. The current work introduces 1) a reasoning process through the processing–structure–property–performance chain and 2) a tool for benchmarking knowledge of LLMs concerning ...
Adrian Ehrenhofer, Thomas Wallmersperger, Gianaurelio Cuniberti   +2 more
wiley   +1 more source

Comparison between parameter-efficient techniques and full fine-tuning: A case study on multilingual news article classification.

PLoS ONE
Adapters and Low-Rank Adaptation (LoRA) are parameter-efficient fine-tuning techniques designed to make the training of language models more efficient. Previous results demonstrated that these methods can even improve performance on some classification ...
Olesya Razuvayevskaya, Ben Wu, João A Leite, Freddy Heppell, Ivan Srba, Carolina Scarton, Kalina Bontcheva, Xingyi Song   +7 more
doaj   +1 more source

Parameter-Efficient Fine-Tuning with Discrete Fourier Transform

CoRR
Accepted by ICML ...
Ziqi Gao, Qichao Wang, Aochuan Chen, Zijing Liu, Bingzhe Wu, Liang Chen 0001, Jia Li 0009   +6 more
openaire   +3 more sources

A Workflow to Accelerate Microstructure‐Sensitive Fatigue Life Predictions

Advanced Engineering Materials, EarlyView.
This study introduces a workflow to accelerate predictions of microstructure‐sensitive fatigue life. Results from frameworks with varying levels of simplification are benchmarked against published reference results. The analysis reveals a trade‐off between accuracy and model complexity, offering researchers a practical guide for selecting the optimal ...
Luca Loiodice, Krzysztof S. Stopka, Michael D. Sangid   +2 more
wiley   +1 more source