no code implementations • NAACL (TrustNLP) 2022 • Kathleen C. Fraser, Svetlana Kiritchenko, Esma Balkir
In an effort to guarantee that machine learning model outputs conform with human moral values, recent work has begun exploring the possibility of explicitly training models to learn the difference between right and wrong.
1 code implementation • 4 Jul 2023 • Isar Nejadgholi, Svetlana Kiritchenko, Kathleen C. Fraser, Esma Balkir
Classifiers tend to learn a false causal relationship between an over-represented concept and a label, which can result in over-reliance on the concept and compromised classification accuracy.
no code implementations • 22 May 2023 • Seraphina Goldfarb-Tarrant, Eddie Ungless, Esma Balkir, Su Lin Blodgett
Bias research in NLP seeks to analyse models for social biases, thus helping NLP practitioners uncover, measure, and mitigate social harms.
1 code implementation • 19 Oct 2022 • Isar Nejadgholi, Esma Balkir, Kathleen C. Fraser, Svetlana Kiritchenko
For a multi-class toxic language classifier, we leverage a concept-based explanation framework to calculate the sensitivity of the model to the concept of sentiment, which has been used before as a salient feature for toxic language detection.
no code implementations • NAACL (TrustNLP) 2022 • Esma Balkir, Svetlana Kiritchenko, Isar Nejadgholi, Kathleen C. Fraser
In this paper, we briefly review trends in explainability and fairness in NLP research, identify the current practices in which explainability methods are applied to detect and mitigate bias, and investigate the barriers preventing XAI methods from being used more widely in tackling fairness issues.
Explainable artificial intelligence Explainable Artificial Intelligence (XAI) +1
no code implementations • 25 May 2022 • Kathleen C. Fraser, Svetlana Kiritchenko, Esma Balkir
In an effort to guarantee that machine learning model outputs conform with human moral values, recent work has begun exploring the possibility of explicitly training models to learn the difference between right and wrong.
1 code implementation • NAACL 2022 • Esma Balkir, Isar Nejadgholi, Kathleen C. Fraser, Svetlana Kiritchenko
We present a novel feature attribution method for explaining text classifiers, and analyze it in the context of hate speech detection.
no code implementations • WS 2020 • Esma Balkir, Daniel Gildea, Shay Cohen
Semiring parsing is an elegant framework for describing parsers by using semiring weighted logic programs.
no code implementations • IJCNLP 2019 • Esma Balkir, Masha Naslidnyk, Dave Palfrey, Arpit Mittal
Bilinear models such as DistMult and ComplEx are effective methods for knowledge graph (KG) completion.
Ranked #18 on Link Prediction on FB15k
no code implementations • 14 Dec 2015 • Esma Balkir, Dimitri Kartsaklis, Mehrnoosh Sadrzadeh
In categorical compositional distributional semantics, phrase and sentence representations are functions of their grammatical structure and representations of the words therein.
no code implementations • 22 Jun 2015 • Esma Balkir, Mehrnoosh Sadrzadeh, Bob Coecke
Categorical compositional distributional model of Coecke et al. (2010) suggests a way to combine grammatical composition of the formal, type logical models with the corpus based, empirical word representations of distributional semantics.