| Full text | |
| Author(s): |
Jodas, Danilo Samuel
;
Garcia, Gabriel Lino
;
Paiola, Pedro Henrique
;
Ribeiro Manesco, Joao Renato
;
Papa, Joao Paulo
Total Authors: 5
|
| Document type: | Journal article |
| Source: | PROGRESS IN PATTERN RECOGNITION, IMAGE ANALYSIS, COMPUTER VISION, AND APPLICATIONS, CIARP 2024, PT I; v. 15368, p. 15-pg., 2025-01-01. |
| Abstract | |
Large Language Models have emerged as transformative agents in the frequently evolving landscape of artificial intelligence, reshaping the world towards a disruptive and modern technological era. This paradigm stresses their crucial role in extending the generative capabilities in the context of natural language processing. Generative Artificial Intelligence, an innovative and cutting-edge research topic, is critical to unlocking remarkable opportunities in our era of unparalleled technological progress. Despite the remarkable progress made in language model architectures, their exponential growth still raises pertinent concerns regarding their deployment and the associated costs for retraining efforts tailored to specific tasks. We present a study achieving a detailed analysis of the impact resulting from the application of diverse quantization methodologies on an open-source large language model tailored for Portuguese classification tasks, aka Bode. Our research thoroughly evaluates the performance nuances introduced by various quantization strategies, thus providing valuable insights into the constant concerns surrounding the optimization of large language models, aiming for enhanced efficiency and effectiveness in growing applications for the Portuguese community. (AU) | |
| FAPESP's process: | 13/07375-0 - CeMEAI - Center for Mathematical Sciences Applied to Industry |
| Grantee: | Francisco Louzada Neto |
| Support Opportunities: | Research Grants - Research, Innovation and Dissemination Centers - RIDC |
| FAPESP's process: | 23/14427-8 - Data Science for Smart Industry (CDII) |
| Grantee: | José Alberto Cuminato |
| Support Opportunities: | Research Grants - Applied Research Centers Program |
| FAPESP's process: | 19/07665-4 - Center for Artificial Intelligence |
| Grantee: | Fabio Gagliardi Cozman |
| Support Opportunities: | Research Grants - Research Program in eScience and Data Science - Research Centers in Engineering Program |
| FAPESP's process: | 23/01374-3 - On the study and development of biological plausible computational intelligent models |
| Grantee: | João Paulo Papa |
| Support Opportunities: | Regular Research Grants |
| FAPESP's process: | 23/03726-4 - On the Study and Development of Multi-method Multi-objective Algorithms |
| Grantee: | Douglas Rodrigues |
| Support Opportunities: | Scholarships in Brazil - Post-Doctoral |
| FAPESP's process: | 23/10823-6 - On the Study and Development of Biological Plausible Computational Intelligent Models |
| Grantee: | Leandro Aparecido Passos Junior |
| Support Opportunities: | Scholarships in Brazil - Support Program for Fixating Young Doctors |
| FAPESP's process: | 24/00789-8 - Domain Invariant Detection of Medical Devices in Plain Chest X-ray Images |
| Grantee: | João Renato Ribeiro Manesco |
| Support Opportunities: | Scholarships in Brazil - Doctorate |