Search Results for author:
Found 18 papers, 8 papers with code
Measuring Fairness with Biased Rulers: A Comparative Study on Bias Metrics for Pre-trained Language Models
We survey the literature on fairness metrics for pre-trained language models and experimentally evaluate compatibility, including both biases in language models and in their downstream tasks.
Addressing the Regulatory Gap: Moving Towards an EU AI Audit Ecosystem Beyond the AIA by Including Civil Society
By considering the value of third-party audits and third-party data access in an audit ecosystem, we identify a regulatory gap in that the Artificial Intelligence Act does not provide access to data for researchers and civil society.
Tik-to-Tok: Translating Language Models One Token at a Time: An Embedding Initialization Strategy for Efficient Language Adaptation
This one-to-many token mapping improves tremendously the initialization of the embedding table for the target language.
Bias, diversity, and challenges to fairness in classification and automated text analysis. From libraries to AI and back
Libraries are increasingly relying on computational methods, including methods from Artificial Intelligence (AI).
Domain Adaptive Decision Trees: Implications for Accuracy and Fairness
In uses of pre-trained machine learning models, it is a known issue that the target population in which the model is being deployed may not have been reflected in the source population with which the model was trained.
How Far Can It Go?: On Intrinsic Gender Bias Mitigation for Text Classification
Considering that the end use of these language models is for downstream tasks like text classification, it is important to understand how these intrinsic bias mitigation strategies actually translate to fairness in downstream tasks and the extent of this.
Political representation bias in DBpedia and Wikidata as a challenge for downstream processing
We describe a case study of the relative over- or under-representation of Belgian political parties between 1990 and 2020 in the English-language DBpedia, the Dutch-language DBpedia, and Wikidata, and highlight the many decisions needed with regard to the design of this data analysis and the assumptions behind it, as well as implications from the results.
RobBERT-2022: Updating a Dutch Language Model to Account for Evolving Language Use
To evaluate if our new model is a plug-in replacement for RobBERT, we introduce two additional criteria based on concept drift of existing tokens and alignment for novel tokens. We found that for certain language tasks this update results in a significant performance increase.
FairDistillation: Mitigating Stereotyping in Language Models
Large pre-trained language models are successfully being used in a variety of tasks, across many languages.
RobBERTje: a Distilled Dutch BERT Model
We found that the performance of the models using the shuffled versus non-shuffled datasets is similar for most tasks and that randomly merging subsequent sentences in a corpus creates models that train faster and perform better on tasks with long sequences.
Measuring Fairness with Biased Rulers: A Survey on Quantifying Biases in Pretrained Language Models
We survey the existing literature on fairness metrics for pretrained language models and experimentally evaluate compatibility, including both biases in language models as in their downstream tasks.
Whistleblower protection in the digital age -- why 'anonymous' is not enough. From technology to a wider view of governance
When technology enters applications and processes with a long tradition of controversial societal debate, multi-faceted new ethical and legal questions arise.
Measuring Shifts in Attitudes Towards COVID-19 Measures in Belgium Using Multilingual BERT
We classify seven months' worth of Belgian COVID-related Tweets using multilingual BERT and relate them to their governments' COVID measures.
Ethical Adversaries: Towards Mitigating Unfairness with Adversarial Machine Learning
These new examples are then used to retrain and improve the model in the first step.
RobBERT: a Dutch RoBERTa-based Language Model
Training a Dutch BERT model thus has a lot of potential for a wide range of Dutch NLP tasks.
Ranked #1 on
Sentiment Analysis
on DBRD
Time to Take Emoji Seriously: They Vastly Improve Casual Conversational Models
Graphical emoji are ubiquitous in modern-day online conversations.
Computational Ad Hominem Detection
Fallacies like the personal attack{---}also known as the ad hominem attack{---}are introduced in debates as an easy win, even though they provide no rhetorical contribution.
AI for the Common Good?! Pitfalls, challenges, and Ethics Pen-Testing
But what is the Common Good, and is it enough to want to be good?
