Explainable Machine Learning

We are interested in studying method to explain relationships between inputs and outputs of black-box machine learning models, particularly in the context of challenging NLU tasks such as fact checking.

We are researching methods for explainable fact checking as part of an ERC Starting Grant project, as well as methods for human-centered explainable retrieval-augmented LLMs as part of a DFF AI project.

Moreover, we are investigating fair and accountable Natural Language Processing methods to understand what influences the employer images that organisations project in job ads, as part of a Carlsberg-funded project.

nlu

Publications

Natural Language Explanations (NLEs) describe how Large Language Models (LLMs) make decisions, drawing on both external Context …
PDF Explainability Project

Multilingual large language models (MLLMs) demonstrate state-of-the-art capabilities across diverse cross-lingual and multilingual …
PDF Limited Data Project Explainability Project

Modern Question Answering (QA) and Reasoning approaches based on Large Language Models (LLMs) commonly use prompting techniques, such …
PDF Explainability Project Knowledge Bases Project

Natural language explanations (NLEs) are commonly used to provide plausible free-text explanations of a model’s reasoning about …
PDF Explainability Project Fact Checking Project

Automated frame analysis of political communication is a popular task in computational social science that is used to study how authors …
PDF Limited Data Project Explainability Project Fact Checking Project

With the rapid development of large language models (LLMs), natural language explanations (NLEs) have become increasingly important for …
PDF Explainability Project

Large language models (LLMs) are capable of generating coherent summaries from very long contexts given a user query. Extracting and …
PDF Limited Data Project Explainability Project

Understanding biases and stereotypes encoded in the weights of Large Language Models (LLMs) is crucial for developing effective …
PDF Bias Detection Project Explainability Project

The growing deployment of large language models (LLMs) across diverse cultural contexts necessitates a better understanding of how the …
PDF Explainability Project Multilingual Project Bias Detection Project

The pervasiveness of large language models and generative AI in online media has amplified the need for effective automated …
PDF Explainability Project Fact Checking Project

Explaining the decision-making process of machine learning models is crucial for ensuring their reliability and fairness. One popular …
PDF Explainability Project Fact Checking Project

Studying human values is instrumental for cross-cultural research, enabling a better understanding of preferences and behaviour of …
PDF Explainability Project Fact Checking Project

We are exposed to much information trying to influence us, such as teaser messages, debates, politically framed news, and propaganda - …
PDF Explainability Project Fact Checking Project

Knowledge-intensive language understanding tasks require Language Models (LMs) to integrate relevant context, mitigating their inherent …
PDF Explainability Project Fact Checking Project

Uncovering latent values and opinions in large language models (LLMs) can help identify biases and mitigate potential harm. Recently, …
PDF Explainability Project Fact Checking Project

The large and ever-increasing amount of data available on the Internet coupled with the laborious task of manual claim and fact …
PDF Explainability Project Fact Checking Project

The emergence of tools based on large language models (LLMs), like OpenAI’s ChatGPT and Google’s Gemini, has garnered immense public …
PDF Fact Checking Project Explainability Project

How much meaning influences gender assignment across languages is an active area of research in modern linguistics and cognitive …
PDF Bias Detection Project Explainability Project

Language Models (LMs) acquire parametric knowledge from their training process, embedding it within their weights. The increasing …
PDF Explainability Project Fact Checking Project

What can large language models learn? By definition, language models (LM) are distributions over strings. Therefore, an intuitive way …
PDF Explainability Project

Explainable AI methods facilitate the understanding of model behaviour, yet, small, imperceptible perturbations to inputs can vastly …
PDF Explainability Project Limited Data Project

Recent studies of the emergent capabilities of transformer-based Natural Language Understanding (NLU) models have indicated that they …
PDF Fact Checking Project Explainability Project

NLP models are used in a variety of critical social computing tasks, such as detecting sexist, racist, or otherwise hateful content. …
PDF Explainability Project Bias Detection Project

Reasoning over spans of tokens from different parts of the input is essential for natural language understanding (NLU) tasks such as …
PDF Explainability Project Fact Checking Project

Answering complex queries on incomplete knowledge graphs is a challenging task where a model needs to answer complex logical queries in …
PDF Knowledge Bases Project Question Answering Project Explainability Project

Explanations of neural models aim to reveal a model’s decision-making process for its predictions. However, recent work shows …
PDF Explainability Project

Language embeds information about social, cultural, and political values people hold. Prior work has explored social and potentially …
PDF Explainability Project Multilingual Project

The success of pre-trained contextualized representations has prompted researchers to analyze them for the presence of linguistic …
PDF Explainability Project Multilingual Project

Fact-checking systems have become important tools to verify fake and misguiding news. These systems become more trustworthy when …
PDF Explainability Project Fact Checking Project Limited Data Project

There have been many efforts to try to understand what grammatical knowledge (e.g., ability to understand the part of speech of a …
PDF Explainability Project Question Answering Project

Two of the most fundamental challenges in Natural Language Understanding (NLU) at present are: (a) how to establish whether deep …
PDF Explainability Project Question Answering Project

With the substantial rise in the amount of mis- and disinformation online, fact checking has become an important task to automate. This …
PDF Fact Checking Project Explainability Project

Counterfactually Augmented Data (CAD) aims to improve out-of-domain generalizability, an indicator of model robustness. The improvement …
PDF Explainability Project Bias Detection Project

The success of multilingual pre-trained models is underpinned by their ability to learn representations shared by multiple languages …
PDF Explainability Project Multilingual Project

Automating the fact checking (FC) process relies on information obtained from external sources. In this work, we posit that it is …
PDF Fact Checking Project Explainability Project

Explanations shed light on a machine learning model’s rationales and can aid in identifying deficiencies in its reasoning …
PDF Explainability Project Limited Data Project

Medical artificial intelligence (AI) systems have been remarkably successful, even outperforming human performance at certain tasks. …
PDF Explainability Project

As NLP models are increasingly deployed in socially situated settings such as online abusive content detection, ensuring these models …
PDF Limited Data Project Explainability Project Bias Detection Project

Sparse attention has been claimed to increase model interpretability under the assumption that it highlights influential inputs. Yet …
PDF Explainability Project

The past decade has seen a substantial rise in the amount of mis- and disinformation online, from targeted disinformation campaigns to …
PDF Fact Checking Project Explainability Project

Recent developments in machine learning have introduced models that approach human performance at the cost of increased architectural …
PDF Code Explainability Project

Adversarial attacks reveal important vulnerabilities and flaws of trained models. One potent type of attack are universal adversarial …
PDF Code Fact Checking Project Explainability Project

While state-of-the-art NLP explainability (XAI) methods focus on supervised, per-instance end or diagnostic probing task evaluation[4, …
PDF Code Limited Data Project Explainability Project

This paper provides the first study of how fact checking explanations can be generated automatically based on available claim context, …
Preprint Fact Checking Project Explainability Project Video