Towards Dialogues for Joint Human-AI Reasoning and Value Alignment