Latest Publications on Prompt Engineering
This is our latest selection of worldwide publications on Prompt Engineering, between many scientific online journals, classified and focused on prompt engineering, prompt design, contextualization, prompt variability, hyperparameter tuning, bias mitigation, zero-shot learning and few-shot learning.
Tip: Further to this selection on Prompt Engineering, you can search and filter our > publication database < by author, topic, keywords, date or journal.
A Novel Approach Assessed in the Caribbean Sea
On 2025-02-02 by David Francisco Bustos Usta, Lien Rodríguez-López, Rafael Ricardo Torres Parra, Luc Bourrel @MDPI
Keywords: forecasting, sea surface temperature, Chronos, upwelling, foundational models
AI summary: Novel approach evaluated in Caribbean Sea. Models compared for SST forecasting. Chronos outperforms Lag-Llama. Highlights benefits and limitations of foundational models for accurate predictions. Benchmark established for future research.
Abstract: Sea surface temperature (SST) plays a pivotal role in air&ndash;sea interactions, with implications for climate, weather, and marine ecosystems, particularly in regions like the Caribbean Sea, where upwelling and dynamic oceanographic processes significantly influence biodiversity and fisheries. This study evaluates the performance of foundational models, Chronos and Lag-Llama, in forecasting SST using 22 years (2002&ndash;2023) of high-resolution satellite-derived and in situ da[...]
The Progress and Prospects of Data Capital for Zero-Shot Deep Brain–Computer Interfaces
On 2025-01-26 by Wenbao Ma, Teng Ma, Daniel Organisciak, Jude E. T. Waide, Xiangxin Meng, Yang Long @MDPI
Keywords: data capital, deep learning, brain-computer interfaces, zero-shot learning, scalability
AI summary: Progress in deep learning for brain-computer interfaces relies on quality data. Zero-shot learning enhances flexibility. Exploration of data capital for large-scale BCI models. Emphasis on paradigm shift with ZSL for BCI technical potential.
Abstract: The vigorous development of deep learning (DL) has been propelled by big data and high-performance computing. For brain&ndash;computer interfaces (BCIs) to benefit from DL in a reliable and scalable manner, the scale and quality of data are crucial. Special emphasis is placed on the zero-shot learning (ZSL) paradigm, which is essential for enhancing the flexibility and scalability of BCI systems. ZSL enables models to generalise from limited examples to new, unseen tasks, addressing data[...]
Enhancing Zero-Shot Learning Through Kernelized Visual Prototypes and Similarity Learning
On 2025-01-26 by Kanglong Cheng, Bowen Fang @MDPI
Keywords: Zero-shot learning, Kernelized visual prototypes, Similarity learning, Domain-shift, Hubness
AI summary: Enhancing ZSL with kernelized prototypes and similarity learning, Addressing challenges with kernelized ridge regression, Introducing kernel polarization and autoencoder structures, Outperforming state-of-the-art methods.
Abstract: Zero-shot learning (ZSL) holds significant promise for scaling image classification to previously unseen classes by leveraging previously acquired knowledge. However, conventional ZSL methods face challenges such as domain-shift and hubness problems. To address these issues, we propose a novel kernelized similarity learning approach that reduces intraclass similarity while increasing interclass similarity. Specifically, we utilize kernelized ridge regression to learn visual prototypes for unseen[...]
Assessing the Sports Understanding Capabilities of Language Models Through Question Answering from Text to Video
On 2025-01-23 by Zhengbang Yang, Haotian Xia, Jingxi Li, Zezhi Chen, Zhuangdi Zhu, Weining Shen @MDPI
Keywords: sports understanding, language models, question answering, NLP, benchmarking
AI summary: Assessing sports understanding capabilities of language models through question answering from text to video. Evaluating mainstream and emerging large models on various sports tasks. Highlighting critical challenges of sports understanding for NLP and future research priorities.
Abstract: Understanding sports presents a fascinating challenge for Natural Language Processing (NLP) due to its intricate and ever-changing nature. Current NLP technologies struggle with the advanced cognitive demands required to reason over complex sports scenarios. To explore the current boundaries of this field, we extensively evaluated mainstream and emerging large models on various sports tasks and addressed the limitations of previous benchmarks. Our study ranges from answering simple queries about[...]
Evaluation of Different Few-Shot Learning Methods in the Plant Disease Classification Domain
On 2025-01-19 by Alexander Uzhinskiy @MDPI
Keywords: few-shot learning, plant disease classification, neural network architectures, similarity learning, loss functions
AI summary: Early detection of plant diseases is crucial. Various methods are evaluated in plant disease classification using neural networks. Cosine-based similarity learning proves superior over Siamese networks. Model organization, training, and data normalization impact generalization ability.
Abstract: Early detection of plant diseases is crucial for agro-holdings, farmers, and smallholders. Various neural network architectures and training methods have been employed to identify optimal solutions for plant disease classification. However, research applying one-shot or few-shot learning approaches, based on similarity determination, to the plantdisease classification domain remains limited. This study evaluates different loss functions used in similarity learning, including Contrastive, Triplet[...]
A Convolutional Neural Network for Early Supraventricular Arrhythmia Identification
On 2025-01-08 by Emilio J. Ochoa, Luis C. Revilla @MDPI
Keywords: convolutional neural network, supraventricular arrhythmia, early identification, ECG signals, deep learning
AI summary: An innovative CNN approach for early SVE detection using ECG signals. Dataset from MIT-BIH database. High precision and recall rates achieved. Significant contribution to cardiac health monitoring.
Abstract: Supraventricular arrhythmias (SVAs), including the often-asymptomatic supraventricular extrasystole (SVE), pose significant challenges in early detection and precise diagnosis. These challenges are of paramount importance, as recurrent SVEs may elevate the risk of developing severe SVAs, potentially resulting in cardiac weakening and subsequent heart failure. In the study conducted, an innovative approach was introduced that combined a convolutional neural network (CNN) architecture to enable th[...]
