Results 151 to 160 of about 227,550 (285)

Fine-Tuning the Qwen2.5-VL Model for Intelligent Applications in the Electrical Domain

EAI Endorsed Transactions on Energy Web
This study explores the fine-tuning application of the Qwen2.5-VL multi modal large model in the electrical domain. The electrical industry faces numerous challenges in maintaining and managing complex electrical systems.
Yao Song, Chunli Lv, Kun Zhu, Xiaobin Qiu   +3 more
doaj   +1 more source

Adaptive Parameter-Efficient Federated Fine-Tuning on Heterogeneous Devices

IEEE Transactions on Mobile Computing
Federated fine-tuning (FedFT) has been proposed to fine-tune the pre-trained language models in a distributed manner. However, there are two critical challenges for efficient FedFT in practical applications, i.e., resource constraints and system heterogeneity.
Jun Liu, Yunming Liao, Hongli Xu, Yang Xu, Jianchun Liu, Chen Qian   +5 more
openaire   +2 more sources

Adaptive 4D‐Printed Vascular Stents With Low‐Temperature‐Activated and Intelligent Deployment

Advanced Functional Materials, EarlyView.
Microarchitected coronary artery stents were fabricated using a polycaprolactone (PCL)‐based shape memory polymer (SMP) composite via projection micro‐stereolithography (PµSL) 4D printing. By incorporating diethyl phthalate (DEP) as a plasticizer, the thermal transition temperature (Ttran) was modulated to about 37°C, enabling rapid and autonomous ...
Yannan Li, Yifan Pan, Chikahiro Imashiro, Chaolun Xu, Jianxian He, Jingao Xu, Kewei Song, Ze Zhang, Chen Gao, Junbo Jiang, Runhuai Yang, Kayo Hirose, Shinjiro Umezu   +12 more
wiley   +1 more source

GateRA: Token-Aware Modulation for Parameter-Efficient Fine-Tuning

Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, DoRA, and HiRA, enable lightweight adaptation of large pre-trained models via low-rank updates. However, existing PEFT approaches apply static, input-agnostic updates to all tokens, disregarding the varying importance and difficulty of different inputs.
Ou, Jie, Jiang, Shuaihong, Du, Yingjun, Snoek, Cees G. M.   +3 more
openaire   +2 more sources

Circular‐Polarization‐Sensitive Organic Photodetectors with a Chiral Nanopatterned Electrode Inverse‐Designed by Genetic Algorithm

Advanced Functional Materials, EarlyView.
A chiral photodetector capable of selectively distinguishing left‐ and right‐handed circularly polarized light is experimentally demonstrated. The device, which features a nanopatterned electrode inverse‐designed by a genetic algorithm within a metal–dielectric–metal nanocavity that incorporates a vacuum‐deposited small‐molecule multilayer, exhibits ...
Kyung Ryoul Park, Jewoong Lee, In‐Suk Choi, Changsoon Kim   +3 more
wiley   +1 more source

Stoma‐Shell Nanoarchitecture for Enhanced Plasma Confinement Catalysis in Synthesis of Ethanol from CO2

Advanced Functional Materials, EarlyView.
High‐energy electron impact in plasma catalysis often causes excessive dissociation of active intermediates, limiting C2+ product selectivity. To address this challenge, a bio‐inspired stoma‐shell nanoarchitecture is designed to decouple electron impact from catalytic reaction zones.
Nan Zou, Zhiliang Dong, Tsun‐Kong Sham, Xiaonian Li, Ying Zheng, Ting Qiu   +5 more
wiley   +1 more source

Certified PEFTSmoothing: Parameter-Efficient Fine-Tuning with Randomized Smoothing

Randomized smoothing is the primary certified robustness method for accessing the robustness of deep learning models to adversarial perturbations in the l2-norm, by adding isotropic Gaussian noise to the input image and returning the majority votes over the base classifier.
Fu, Chengyan, Wang, Wenjie
openaire   +2 more sources

Photoconductivity‐Driven Quantum Efficiency Gain in Inorganic Ruddlesden‐Popper Layered Cs2PbBr2I2 Perovskite Photodetector for Visible Light Detection

Advanced Functional Materials, EarlyView.
Rational halogen mixing strategy was employed to shift the bandgap of Cs2PbBr2I2 from ultraviolet to visible region, enabling first realization of a visible‐light photodetector with this 2D layered Ruddlesden‐Popper perovskite material. Under illumination, light‐induced internal field forms and drives trap‐mediated persistent photoconductivity ...
Md Fahim Al Fattah, Asif Abdullah Khan, Saikiran Khamgaonkar, Majid Haji Bagheri, Danyah Khan, Md Rasidul Islam, Mahmoud Almadoun, Aixi Pan, Alexander Li, Marwa Abd‐Ellah, Vivek Maheshwari, Dayan Ban   +11 more
wiley   +1 more source

Lithium Intercalation in the Anisotropic Van Der Waals Semiconductor CrSBr

Advanced Functional Materials, EarlyView.
We report the lithium intercalation in the layered van der Waals crystal CrSBr, revealing strongly anisotropic ion‐migration dynamics. Optical and electrical characterization of exfoliated CrSBr shows lithium diffusion coefficients that differ by more than an order of magnitude along a‐ and b‐directions, consistent with molecular dynamics simulations ...
Kseniia Mosina, Aljoscha Söll, Jiří Šturala, Martin Veselý, Petr Levinsky, Ondřej Kaman, Florian Dirnberger, Giuliana Materzanini, Nicola Marzari, Gian‐Marco Rignanese, Borna Radatović, Vlastimil Mazánek, David Sedmidubský, Zdeněk Sofer   +13 more
wiley   +1 more source

PEFT-U: Parameter-Efficient Fine-Tuning for User Personalization

The recent emergence of Large Language Models (LLMs) has heralded a new era of human-AI interaction. These sophisticated models, exemplified by Chat-GPT and its successors, have exhibited remarkable capabilities in language understanding. However, as these LLMs have undergone exponential growth, a crucial dimension that remains understudied is the ...
Clarke, Christopher, Heng, Yuzhao, Tang, Lingjia, Mars, Jason   +3 more
openaire   +2 more sources