2024/04/04 by Makesh Narsimhan Sreedhar, Traian Rebedea, Sreedhar, Makesh Narsimhan +5 · 2 citations
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Speech and dialogue systems
paper · pdf · doi:10.48550/arxiv.2404.03820
openalex publication_date 2024/04/04 · openalex created_date 2024/04/09 · openalex updated_date 2026/07/28
Recent advancements in instruction-tuning datasets have predominantly focused on specific tasks like mathematical or logical reasoning. There has been a notable gap in data designed for aligning language models to maintain topic relevance in conversations - a critical aspect for deploying chatbots to production. We introduce the CantTalkAboutThis dataset to help language models remain focused on the subject at hand during task-oriented interactions. It consists of synthetic dialogues on a wide range of conversation topics from different domains. These dialogues are interspersed with distractor turns that intentionally divert the chatbot from the predefined topic. Fine-tuning language models on this dataset helps make them resilient to deviating from the role assigned and improves their ability to maintain topical coherence compared to general-purpose instruction-tuned LLMs like GPT-4-turbo and Mixtral-Instruct. Additionally, preliminary observations suggest that training models on this dataset also enhance their performance on fine-grained instruction following tasks, including safety alignment.