Results 11 to 20 of about 2,403,325 (304)
Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning [PDF]
Neural Information Processing Systems, 2022 Few-shot in-context learning (ICL) enables pre-trained language models to perform a previously-unseen task without any gradient-based training by feeding a small number of training examples as part of the input.
Haokun Liu +6 more
semanticscholar +1 more source
BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models [PDF]
Annual Meeting of the Association for Computational Linguistics, 2021 We introduce BitFit, a sparse-finetuning method where only the bias-terms of the model (or a subset of them) are being modified. We show that with small-to-medium training data, applying BitFit on pre-trained BERT models is competitive with (and ...
Elad Ben-Zaken +2 more
semanticscholar +1 more source
Journal of Marine Science and Engineering, 2021 A method based on a coarse- and fine-tuning fixed-grid wavelet networks is presented for online prediction of the coupled heave-pitch motions of a ship in irregular waves.
Baigang Huang +2 more
doaj +1 more source
Is The Fine-Tuning Argument Theistic? [PDF]
پژوهشنامه فلسفه دین, 2021 I intend to show that the fine-tuning argument is not a theistic one, and is therefore not desirable for the theist. I will argue that (1) if the fine-tuning of the cosmos exceeds that which is required for the existence of human beings, it would not be ...
Mohammad Ebrahim Maqsudi
doaj +1 more source
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention [PDF]
arXiv.org, 2023 We present LLaMA-Adapter, a lightweight adaption method to efficiently fine-tune LLaMA into an instruction-following model. Using 52K self-instruct demonstrations, LLaMA-Adapter only introduces 1.2M learnable parameters upon the frozen LLaMA 7B model ...
Renrui Zhang +8 more
semanticscholar +1 more source
International Journal of Modern Physics A, 1993 We develop further a new geometrical model of a discretized string, proposed in Ref. 1, and establish its basic physical properties. The model can be considered as the natural extension of the usual Feynman amplitude of the random walks to random surfaces.
Savvidy, G. K., Savvidy, K. G.
openaire +2 more sources
Physical Review D, 2018 We introduce a mathematical framework for quantifying fine-tuning in general physical settings. In particular, we identify two distinct perspectives on fine-tuning, namely, a local and a global perspective --- and develop corresponding measures. These measures apply broadly to settings characterized by an arbitrary number of observables whose values ...
Azhar, Feraz, Loeb, Abraham
openaire +2 more sources
Future Internet, 2023 With sentiment prediction technology, businesses can quickly look at user reviews to find ways to improve their products and services. We present the BertBilstm Multiple Emotion Judgment (BBMEJ) model for small-sample emotion prediction tasks to solve ...
Jinlong Wang, Dong Cui, Qiang Zhang
doaj +1 more source
Fine-Tuning Language Models with Just Forward Passes [PDF]
Neural Information Processing Systems, 2023 Fine-tuning language models (LMs) has yielded success on diverse downstream tasks, but as LMs grow in size, backpropagation requires a prohibitively large amount of memory.
Sadhika Malladi +6 more
semanticscholar +1 more source
World Journal of Surgery, 2011 Communication has become a major issue in cancer care, and rightly so. Good communication helps patients and their families understand the situation and assists with decision-making, acceptance, adjustment, and coping. It is well known that the way in which a cancer diagnosis is delivered not only affects the person’s understanding of the illness, it ...
Elisabeth E, Elder, Raimund, Jakesz
openaire +2 more sources