We demonstrate that universal adversarial triggers can control both the topic and stance of GPT-2’s generated text, revealing security vulnerabilities in deployed language models and proposing constructive applications for bias auditing.
We present an unsupervised algorithm for inducing semantic networks from Wiktionary’s crowd-sourced data, creating a WordNet-like resource an order of magnitude larger than Princeton WordNet with over 344,000 linked example sentences.
QuAC introduces a conversational QA dataset that models student-teacher interactions, creating context-dependent questions that test systems’ ability to understand dialogue and resolve references.
CoQA extends question answering beyond isolated questions to conversations that require context and reference understanding.
Learn how computers understand words through mathematical vectors, from simple counting methods to contextual embeddings that power modern NLP.