Picture for Hugo Abonizio

Hugo Abonizio

Sabiá-3 Technical Report

Add code
Oct 15, 2024
Viaarxiv icon

Sabiá-2: A New Generation of Portuguese Large Language Models

Add code
Mar 26, 2024
Viaarxiv icon

Evaluating GPT-4's Vision Capabilities on Brazilian University Admission Exams

Add code
Nov 23, 2023
Viaarxiv icon

InPars Toolkit: A Unified and Reproducible Synthetic Data Generation Pipeline for Neural Information Retrieval

Add code
Jul 10, 2023
Viaarxiv icon

Sabiá: Portuguese Large Language Models

Add code
Apr 18, 2023
Viaarxiv icon

InPars-v2: Large Language Models as Efficient Dataset Generators for Information Retrieval

Add code
Jan 14, 2023
Viaarxiv icon

In Defense of Cross-Encoders for Zero-Shot Retrieval

Add code
Dec 12, 2022
Viaarxiv icon

MonoByte: A Pool of Monolingual Byte-level Language Models

Add code
Sep 27, 2022
Figure 1 for MonoByte: A Pool of Monolingual Byte-level Language Models
Figure 2 for MonoByte: A Pool of Monolingual Byte-level Language Models
Figure 3 for MonoByte: A Pool of Monolingual Byte-level Language Models
Viaarxiv icon

No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval

Add code
Jun 06, 2022
Figure 1 for No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval
Figure 2 for No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval
Figure 3 for No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval
Figure 4 for No Parameter Left Behind: How Distillation and Model Size Affect Zero-Shot Retrieval
Viaarxiv icon

Billions of Parameters Are Worth More Than In-domain Training Data: A case study in the Legal Case Entailment Task

Add code
May 30, 2022
Figure 1 for Billions of Parameters Are Worth More Than In-domain Training Data: A case study in the Legal Case Entailment Task
Figure 2 for Billions of Parameters Are Worth More Than In-domain Training Data: A case study in the Legal Case Entailment Task
Figure 3 for Billions of Parameters Are Worth More Than In-domain Training Data: A case study in the Legal Case Entailment Task
Viaarxiv icon