Papers with f-score

6 papers
Leveraging Intra-User and Inter-User Representation Learning for Automated Hate Speech Detection (N18-2)

Copied to clipboard

Challenge: Existing methods that focus on a single tweet as input are likely to yield high false positive and negative rates.
Approach: They propose a model that leverages intra-user and inter-user representation learning to improve hate speech detection on Twitter by suppressing the noise in a single Tweet.
Outcome: The proposed model significantly improves the f-score of a strong bidirectional LSTM model by 10.1%.
Neural Ranking Models for Temporal Dependency Structure Parsing (D18-1)

Copied to clipboard

Challenge: a new neural temporal dependency parser is being developed for news reports and narrative stories . a similar system is used for other NLP applications such as timeline construction .
Approach: They build a neural temporal dependency parser that parses time expressions and events in a text . their results shed light on the nature of temporal relation structures in different domains .
Outcome: The proposed model beats baselines on news reports and narrative stories on two data domains.
Disfluency Detection using Auto-Correlational Neural Networks (D18-1)

Copied to clipboard

Challenge: a recent study proposes an auto-correlational neural network (ACNN) that can detect disfluency in speech . the model uses a convolutional neural system and augments it with a new auto-corrector .
Approach: They propose a convolutional neural network model that captures "rough copy" dependencies . the model is based on a new auto-correlation operator that capture the kinds of "rough copies" dependency .
Outcome: The proposed model outperforms the baseline CNN on a disfluency detection task with a 5% increase in f-score.
Implementation and Evaluation of an LFG-based Parser for Wolof (2020.lrec-1)

Copied to clipboard

Challenge: a parsing system for Wolof is developed based on the Lexical Functional Grammar (LFG) system provides detailed syntactic analysis essential for the further development of NLP applications.
Approach: They propose a parsing system for Wolof based on the Lexical Functional Grammar (LFG) system uses finite-state transducers for word tokenization and morphological analysis .
Outcome: The proposed system achieves 67.2% recall, 92.8% precision and an f-score of 77.9%.
Logic-driven Indirect Supervision: An Application to Crisis Counseling (2023.acl-long)

Copied to clipboard

Challenge: Text-based crisis counseling services are increasingly adopted by people seeking confidential mental health support.
Approach: They propose an inexpensive method that exploits declaratively stated structural dependencies between both levels of annotation to improve utterance modeling.
Outcome: The proposed method improves utterance modeling by 3.5% over a strong multitask baseline.
Geographically-Informed Language Identification (2024.lrec-main)

Copied to clipboard

Challenge: a paper develops a method to identify languages based on geographic origin of text . the model is based in regions where languages are widely spoken and may occur anywhere .
Approach: They propose to incorporate geographic information into a language identification model to ensure coverage of linguae francae regardless of location.
Outcome: The proposed model includes 31 widely-spoken international languages . the model improves on social media data and improves performance on 916 languages compared to baseline models .

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations