相关论文: Quantifying and Attributing Polarization to Annota…

Warning: this paper contains content that may be offensive or upsetting Hate speech moderation on global platforms poses unique challenges due to the multimodal and multilingual nature of content, along with the varying cultural…

Stance detection is nearly always formulated as classifying text into Favor, Against, or Neutral. This convention was inherited from debate analysis and has been applied without modification to social media since SemEval-2016. However,…

For subjective tasks such as hate detection, where people perceive hate differently, the Large Language Model's (LLM) ability to represent diverse groups is unclear. By including additional context in prompts, we comprehensively analyze…

Many NLP tasks exhibit human label variation, where different annotators give different labels to the same texts. This variation is known to depend, at least in part, on the sociodemographics of annotators. Recent research aims to model…

Current alignment pipelines presume a single, universal notion of desirable behavior. However, human preferences often diverge across users, contexts, and cultures. As a result, disagreement collapses into the majority signal and minority…

When annotators disagree on a label, the disagreement itself carries signal -- and the number of annotators needed to capture it depends on the evaluation metric. We fine-tune NLI models on label distributions subsampled from ChaosNLI, a…

Aggregating multiple annotations into a single ground truth label may hide valuable insights into annotator disagreement, particularly in tasks where subjectivity plays a crucial role. In this work, we explore methods for identifying…

We propose an annotation approach that captures not only labels but also the reading process underlying annotators' decisions, e.g., what parts of the text they focus on, re-read or skim. Using this framework, we conduct a case study on the…

Content moderation and toxicity classification represent critical tasks with significant social implications. However, studies have shown that major classification models exhibit tendencies to magnify or reduce biases and potentially…

With the increasing diversity of use cases of large language models, a more informative treatment of texts seems necessary. An argumentative analysis could foster a more reasoned usage of chatbots, text completion mechanisms or other…

Human-annotated preference data play an important role in aligning large language models (LLMs). In this paper, we study two connected questions: how to monitor the quality of human preference annotators and how to incentivize them to…

We present a novel feature attribution method for explaining text classifiers, and analyze it in the context of hate speech detection. Although feature attribution models usually provide a single importance score for each token, we instead…

Human annotation remains the foundation of reliable and interpretable data in Natural Language Processing (NLP). As annotation and evaluation tasks continue to expand, from categorical labelling to segmentation, subjective judgment, and…

Hateful speech detection is a key component of content moderation, yet current evaluation frameworks rarely assess why a text is deemed hateful. We introduce \textsf{HateXScore}, a four-component metric suite designed to evaluate the…

Detecting problematic content, such as hate speech, is a multifaceted and ever-changing task, influenced by social dynamics, user populations, diversity of sources, and evolving language. There has been significant efforts, both in academia…

Hate speech classification has become an important problem due to the spread of hate speech on social media platforms. For a given set of Hate Terms lists (HTs-lists) and Hate Speech data (HS-data), it is challenging to understand which…

Toxicity is an increasingly common and severe issue in online spaces. Consequently, a rich line of machine learning research over the past decade has focused on computationally detecting and mitigating online toxicity. These efforts…

Hate speech (HS) classifiers do not perform equally well in detecting hateful expressions towards different target identities. They also demonstrate systematic biases in predicted hatefulness scores. Tapping on two recently proposed…

Machine learning models are commonly used to detect toxicity in online conversations. These models are trained on datasets annotated by human raters. We explore how raters' self-described identities impact how they annotate toxicity in…

Annotator disagreement is ubiquitous in natural language processing (NLP) tasks. There are multiple reasons for such disagreements, including the subjectivity of the task, difficult cases, unclear guidelines, and so on. Rather than simply…