Search Results for author: Alvaro Soto

Found 32 papers, 8 papers with code

Behind the Magic, MERLIM: Multi-modal Evaluation Benchmark for Large Image-Language Models

1 code implementation3 Dec 2023 Andrés Villa, Juan Carlos León Alcázar, Alvaro Soto, Bernard Ghanem

This paper introduces a Multi-modal Evaluation Benchmark named MERLIM, a scalable test-bed to assess the performance of IT-LVLMs on fundamental computer vision tasks.


Studying Generalization on Memory-Based Methods in Continual Learning

no code implementations16 Jun 2023 Felipe del Rio, Julio Hurtado, Cristian Buc, Alvaro Soto, Vincenzo Lomonaco

One of the objectives of Continual Learning is to learn new concepts continually over a stream of experiences and at the same time avoid catastrophic forgetting.

Continual Learning Out-of-Distribution Generalization

A Memory Model for Question Answering from Streaming Data Supported by Rehearsal and Anticipation of Coreference Information

no code implementations12 May 2023 Vladimir Araujo, Alvaro Soto, Marie-Francine Moens

Existing question answering methods often assume that the input content (e. g., documents or videos) is always accessible to solve the task.

Memorization Question Answering

How Relevant is Selective Memory Population in Lifelong Language Learning?

no code implementations3 Oct 2022 Vladimir Araujo, Helena Balabin, Julio Hurtado, Alvaro Soto, Marie-Francine Moens

Lifelong language learning seeks to have models continuously learn multiple tasks in a sequential order without suffering from catastrophic forgetting.

Question Answering text-classification +1

A Study on the Predictability of Sample Learning Consistency

no code implementations7 Jul 2022 Alain Raymond-Saez, Julio Hurtado, Alvaro Soto

Curriculum Learning is a powerful training method that allows for faster and better training in some settings.

Memory Population in Continual Learning via Outlier Elimination

1 code implementation4 Jul 2022 Julio Hurtado, Alain Raymond-Saez, Vladimir Araujo, Vincenzo Lomonaco, Alvaro Soto, Davide Bacciu

This paper introduces Memory Outlier Elimination (MOE), a method for identifying and eliminating outliers in the memory buffer by choosing samples from label-homogeneous subpopulations.

Continual Learning

Entropy-based Stability-Plasticity for Lifelong Learning

1 code implementation18 Apr 2022 Vladimir Araujo, Julio Hurtado, Alvaro Soto, Marie-Francine Moens

The ability to continuously learn remains elusive for deep learning models.

Augmenting BERT-style Models with Predictive Coding to Improve Discourse-level Representations

no code implementations EMNLP 2021 Vladimir Araujo, Andrés Villa, Marcelo Mendoza, Marie-Francine Moens, Alvaro Soto

Current language models are usually trained using a self-supervised scheme, where the main focus is learning representations at the word or sentence level.

Relationship Detection Sentence

DeepSocNav: Social Navigation by Imitating Human Behaviors

no code implementations19 Jul 2021 Juan Pablo de Vicente, Alvaro Soto

Current datasets to train social behaviors are usually borrowed from surveillance applications that capture visual data from a bird's-eye perspective.

Social Navigation Unity

Optimizing Reusable Knowledge for Continual Learning via Metalearning

1 code implementation NeurIPS 2021 Julio Hurtado, Alain Raymond-Saez, Alvaro Soto

On the other hand, a set of trainable masks provides the key mechanism to selectively choose from the KB relevant weights to solve each task.

Continual Learning

Inspecting the concept knowledge graph encoded by modern language models

no code implementations Findings (ACL) 2021 Carlos Aspillaga, Marcelo Mendoza, Alvaro Soto

The field of natural language understanding has experienced exponential progress in the last few years, with impressive results in several tasks.

Natural Language Understanding Word Embeddings

Tracking the progress of Language Models by extracting their underlying Knowledge Graphs

no code implementations1 Jan 2021 Carlos Aspillaga, Marcelo Mendoza, Alvaro Soto

The state of the art, previously dominated by pre-trained word embeddings, is now being pushed forward by large pre-trained contextual representation models.

Knowledge Graphs Word Embeddings

Catching the Long Tail in Deep Neural Networks

no code implementations1 Jan 2021 Julio Hurtado, Alain Raymond, Alvaro Soto

As a working hypothesis, we speculate that during learning some weights focus on mining patterns from frequent examples while others are in charge of memorizing rare long-tail samples.


Temporal Attention Modules for Memory-Augmented Neural Networks

no code implementations1 Jan 2021 Rodolfo Palma, Alvaro Soto, Luis Martí, Nayat Sanchez-pi

We introduce two temporal attention modules which can be plugged into traditional memory augmented recurrent neural networks to improve their performance in natural language processing tasks.

A Survey on Deep Learning and Explainability for Automatic Report Generation from Medical Images

no code implementations20 Oct 2020 Pablo Messina, Pablo Pino, Denis Parra, Alvaro Soto, Cecilia Besa, Sergio Uribe, Marcelo andía, Cristian Tejos, Claudia Prieto, Daniel Capurro

Every year physicians face an increasing demand of image-based diagnosis from patients, a problem that can be addressed with recent artificial intelligence methods.

Medical Report Generation

Translating Natural Language Instructions for Behavioral Robot Navigation with a Multi-Head Attention Mechanism

no code implementations WS 2020 Patricio Cerda-Mardini, Vladimir Araujo, Alvaro Soto

We propose a multi-head attention mechanism as a blending layer in a neural network model that translates natural language to a high level behavioral language for indoor robot navigation.

Robot Navigation

Differentiable Adaptive Computation Time for Visual Reasoning

1 code implementation CVPR 2020 Cristobal Eyzaguirre, Alvaro Soto

This paper presents a novel attention-based algorithm for achieving adaptive computation called DACT, which, unlike existing ones, is end-to-end differentiable.

Visual Reasoning

Interpretable Visual Question Answering by Visual Grounding from Attention Supervision Mining

no code implementations1 Aug 2018 Yundong Zhang, Juan Carlos Niebles, Alvaro Soto

A key aspect of VQA models that are interpretable is their ability to ground their answers to relevant regions in the image.

Question Answering Visual Grounding +1

Comparing Neural and Attractiveness-based Visual Features for Artwork Recommendation

no code implementations22 Jun 2017 Vicente Dominguez, Pablo Messina, Denis Parra, Domingo Mery, Christoph Trattner, Alvaro Soto

Advances in image processing and computer vision in the latest years have brought about the use of visual features in artwork recommendation.

Exploring Content-based Artwork Recommendation with Metadata and Visual Features

no code implementations19 Jun 2017 Pablo Messina, Vicente Dominguez, Denis Parra, Christoph Trattner, Alvaro Soto

Compared to other areas, artwork recommendation has received little attention, despite the continuous growth of the artwork market.

How a General-Purpose Commonsense Ontology can Improve Performance of Learning-Based Image Retrieval

1 code implementation24 May 2017 Rodrigo Toro Icarte, Jorge A. Baier, Cristian Ruz, Alvaro Soto

Consequently, a main conclusion of this work is that general-purpose commonsense ontologies improve performance on visual reasoning tasks when properly filtered to select meaningful visual relations.

Image Retrieval Retrieval +2

Action Recognition in Video Using Sparse Coding and Relative Features

no code implementations CVPR 2016 Anali Alfaro, Domingo Mery, Alvaro Soto

In terms of the method to obtain key-sequences, we introduce a loss function that, for each video, leads to the identification of a sparse set of representative key-frames capturing both, relevant particularities arising in the input video, as well as relevant generalities arising in the complete class collection.

Action Recognition Temporal Action Localization

Discriminative Hierarchical Modeling of Spatio-Temporally Composable Human Activities

no code implementations CVPR 2014 Ivan Lillo, Alvaro Soto, Juan Carlos Niebles

Our method describes human activities in a hierarchical discriminative model that operates at three semantic levels.

Cannot find the paper you are looking for? You can Submit a new open access paper.