Results 91 to 100 of about 173,284 (257)
LoRMA: Low-Rank Multiplicative Adaptation for LLMs
Findings of the Association for Computational Linguistics: ACL 2025Large Language Models have shown remarkable capabilities in the NLP domain. Their effectiveness can mainly be attributed to their ability to adapt to an array of downstream tasks. However, generally, full fine-tuning is a computationally expensive job.
Harsh Bihany +2 more
openaire +3 more sources
Molecular Oncology, EarlyView.CIN85 is highly expressed in osteosarcoma, particularly in metastatic lesions. Its overexpression increases cell migration and Matrigel invasion, while silencing CIN85 suppresses these behaviors. Transcriptome analysis shows that CIN85 regulates MMP2, COL3A1, and Akt/mTOR signaling. Targeting these pathways reverses CIN85‐induced motility, highlighting
Iryna Horak +10 more
wiley +1 more source
Adaptation to climate in widespread eucalypt species [PDF]
, 2013 The long term success of revegetation efforts will depend upon the planted species’ resilience to climate change. Many widespread species grow across a range of climatic conditions and, thus, may possess adaptations that could be utilised to improve ...
Suzanne Prober +2 more
core
Serial Low-rank Adaptation of Vision Transformer
2025 IEEE International Conference on Multimedia and Expo (ICME)Fine-tuning large pre-trained vision foundation models in a parameter-efficient manner is critical for downstream vision tasks, considering the practical constraints of computational and storage costs. Low-rank adaptation (LoRA) is a well-established technique in this domain, achieving impressive efficiency by reducing the parameter space to a low-rank
Houqiang Zhong +9 more
openaire +2 more sources
Molecular Oncology, EarlyView.Intratumour heterogeneity complicates precision management of advanced endometrial cancer. Circulating tumor DNA (ctDNA) offers a minimally invasive strategy to capture tumor evolution and therapeutic resistance. Here, we compare tumor‐agnostic NGS with tumor‐informed ddPCR, outlining their relative sensitivity, concordance, and clinical implications ...
Carlos Casas‐Arozamena +15 more
wiley +1 more source
Discriminative Transfer Subspace Learning via Low-Rank and Sparse Representation
, 2016 In this paper, we address the problem of unsupervised domain transfer learning in which no labels are available in the target domain. We use a transformation matrix to transfer both the source and target data to a common subspace, where each target ...
Xu, Yong +4 more
core +1 more source
Interpreting the effects of DNA polymerase variants at the structural level
Molecular Oncology, EarlyView.Using MAVISp and molecular dynamics simulations, we analyzed over 60 000 missense variants in POLE and POLD1 from ClinVar, COSMIC, cBioPortal, and saturation mutagenesis. Identified mechanistic indicators, including stability, binding, and long‐range, enable structural interpretation, providing ACMG‐like evidence for possible reclassification of VUS ...
Matteo Arnaudi +7 more
wiley +1 more source
LoTR: Low Tensor Rank Weight Adaptation
In this paper we generalize and extend an idea of low-rank adaptation (LoRA) of large language models (LLMs) based on Transformer architecture. Widely used LoRA-like methods of fine-tuning LLMs are based on matrix factorization of gradient update.
Cherniuk, Daria +4 more
core
Molecular Oncology, EarlyView.Dormant cancer cells can hide in distant organs for years, evading treatment and the immune system. This review highlights how signals from the surrounding tissue and immune environment keep these cells inactive or trigger their reawakening. Understanding these mechanisms may help develop therapies to eliminate or control dormant cells and prevent ...
Kanishka Tiwary +1 more
wiley +1 more source