4 papers
ConspirED: A Dataset for Cognitive Traits of Conspiracy Theories and Large Language Model Safety
Luke Bates, Max Glockner, Preslav Nakov +1
Conspiracy theories erode public trust in science and institutions while resisting debunking by evolving and absorbing counter-evidence. As AI-generated misinformation becomes incr…
NeoQA: Evidence-based Question Answering with Generated News Events
Max Glockner, Xiang Jiang, Leonardo F. R. Ribeiro +2
Evaluating Retrieval-Augmented Generation (RAG) in large language models (LLMs) is challenging because benchmarks can quickly become stale. Questions initially requiring retrieval…
Self-Rationalization in the Wild: A Large Scale Out-of-Distribution Evaluation on NLI-related tasks
Jing Yang, Max Glockner, Anderson Rocha +1
Free-text explanations are expressive and easy to understand, but many datasets lack annotated explanation data, making it challenging to train models for explainable predictions.…
Grounding Fallacies Misrepresenting Scientific Publications in Evidence
Max Glockner, Yufang Hou, Preslav Nakov +1
Health-related misinformation claims often falsely cite a credible biomedical publication as evidence. These publications only superficially seem to support the false claim, when l…