Výsledky vyhledávání - "López, Jose Antonio"

Report

ALPINE: An adaptive language-agnostic pruning method for language models for code

Autor: Saad, Mootez, López, José Antonio Hernández, Chen, Boqi, Varró, Dániel, Sharma, Tushar

Language models of code have demonstrated state-of-the-art performance across various software engineering and source code analysis tasks. However, their demanding computational resource requirements and consequential environmental footprint remain a

Externí odkaz: http://arxiv.org/abs/2407.04147

Zobrazit plný text záznamu

Report

On Inter-dataset Code Duplication and Data Leakage in Large Language Models

Autor: López, José Antonio Hernández, Chen, Boqi, Saaz, Mootez, Sharma, Tushar, Varró, Dániel

Motivation. Large language models (LLMs) have exhibited remarkable proficiency in diverse software engineering (SE) tasks. Handling such tasks typically involves acquiring foundational coding knowledge on large, general-purpose datasets during a pre-

Externí odkaz: http://arxiv.org/abs/2401.07930

Zobrazit plný text záznamu

Report

AST-Probe: Recovering abstract syntax trees from hidden representations of pre-trained language models

Autor: López, José Antonio Hernández, Weyssow, Martin, Cuadrado, Jesús Sánchez, Sahraoui, Houari

The objective of pre-trained language models is to learn contextual representations of textual data. Pre-trained language models have become mainstream in natural language processing and code modeling. Using probes, a technique to study the linguisti

Externí odkaz: http://arxiv.org/abs/2206.11719

Zobrazit plný text záznamu