Publications by authors named "F Pieri Nerli"
Article Synopsis
- Large language models (LLMs) like GPT-J-6B, Llama-3.1-8B, and Mistral-7B can learn chemical properties effectively through fine-tuning without specialized features.
- Fine-tuning these models often outperforms traditional machine learning methods in simple classification tasks, with potential success in more complex problems depending on dataset size and question type.
- The ease of converting datasets for LLM training and the effectiveness of small datasets in generating predictive models suggest that LLMs could significantly streamline experimental processes in chemical research.
From January 1986 to February 1993, 85 patients with spontaneous pneumothorax were treated in our department by a new endoscopic procedure using an Nd-YAG laser beam via thoracoscopy to obtain permanent pleurodesis and to treat the lung lesion responsible for the air leak. The 55 men and 30 women ranged in age from 16 to 51 years (mean age 26 years). Under general anesthesia the thoracoscope was introduced through a 1-cm incision in the anterior axillary line of the fourth intercostal space.
View Article and Find Full Text PDFOur experience with 16 patients operated upon for chronic mitral insufficiency secondary to coronary artery disease is reported. The patients ranged from 40 to 65 years in age; all were in class III or IV of the N.Y.
View Article and Find Full Text PDF[Immediate results in 43 cases of atrioventricular canal correction].
Boll Soc Ital Cardiol
December 1975