Papers by Denilson Barbosa
Cree Corpus: A Collection of nêhiyawêwin Resources (2022.acl-long)
Copied to clipboard
| Challenge: | Plains Cree is a low resource language with no corpus available for development . a lack of publicly available corpora hinders the development of such technologies . |
| Approach: | They develop a corpus of Plains Cree (nêhiyawêwin) covering genres, time periods, and texts for a variety of intended audiences. |
| Outcome: | The corpus covers genres, time periods, and texts for a variety of intended audiences. |
Accurate and Nuanced Open-QA Evaluation Through Textual Entailment (2024.findings-acl)
Copied to clipboard
| Challenge: | Open-domain question answering (Open-QA) evaluations are criticized for the ambiguity in questions and the lack of semantic understanding in evaluators. |
| Approach: | They propose to examine the entailment relations of answers to identify more informative and more general system answers. |
| Outcome: | The proposed evaluations offer a much closer evaluation to human judgment on NaturalQuestions and TriviaQA while being learning-free. |
WordTies: Measuring Word Associations in Language Models via Constrained Sampling (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Word associations are widely used in psychology to provide insights on how humans perceive and understand concepts. |
| Approach: | They propose an algorithm that allows an asymmetric measurement of associated words, given a cue word as input. |
| Outcome: | The proposed algorithm shares more overlap with human associations and observes the asymmetric property of human associations. |
Connecting Language and Knowledge with Heterogeneous Representations for Neural Relation Extraction (N19-1)
Copied to clipboard
| Challenge: | Knowledge Bases (KBs) require constant updating to reflect changes to the world they represent. |
| Approach: | They propose a framework that unifies learning of RE and KBE models . the framework is based on a relation extraction task that uses a KB relation to a phrase . |
| Outcome: | The proposed framework unifies learning of RE and KBE models, leading to significant improvements over the state-of-the-art RE framework. |
NLP Workbench: Efficient and Extensible Integration of State-of-the-art Text Mining Tools (2023.eacl-demo)
Copied to clipboard
| Challenge: | NLP Workbench is a web-based text mining platform that allows non-expert users to obtain semantic understanding of large-scale corpora using state-of-the-art text mining models. |
| Approach: | They propose to use a microservice architecture to replace existing models or integrate a new one. |
| Outcome: | The proposed model is extensible and can be easily replaced or integrated with existing models. |
KnowledgeNet: A Benchmark Dataset for Knowledge Base Population (D19-1)
Copied to clipboard
| Challenge: | KnowledgeNet provides text exhaustively annotated with facts . high-quality KBs still rely almost exclusively on human-curated structured or semi-structured data. |
| Approach: | They propose five baseline approaches to populating a knowledge base with facts . the best approach achieves an F1 score of 0.50, significantly outperforming a traditional approach by 79% . |
| Outcome: | The best approach achieves an F1 score of 0.50, outperforming a traditional approach by 79%, indicating the dataset is challenging. |
Neural Fine-Grained Entity Type Classification with Hierarchy-Aware Loss (N18-1)
Copied to clipboard
| Challenge: | Existing methods for fine-grained type classification rely on distant supervision and are susceptible to noisy labels that can be out-of-context or overly-specific. |
| Approach: | They propose a neural network model that uses cross-entropy loss function to handle out-of-context labels and hierarchical loss normalization to cope with overly-specific ones. |
| Outcome: | The proposed model outperforms the state-of-the-art on established benchmarks for the task. |
Improving Bengali and Hindi Large Language Models (2024.lrec-main)
Copied to clipboard
| Challenge: | Bengali and Hindi are low-resource languages, and the state-of-the-art tokenization methods fail to separate roots from affixes. |
| Approach: | They used BERT and Wordpiece tokenizers to train a wordpiece tokenization system for Bengali and Hindi to model fine-grained character-level information. |
| Outcome: | The proposed tokenizers outperform the state-of-the-art and Wordpiece tokenizer for modeling Bengali and Hindi. |