World Modeling Embodied Agents Interactive Narratives Teach agents to operate in complex text-based simulations World Modeling with Language Improving the ability of LLMs to act as world models that can help AI agents plan and execute goals. Multimodal Language Grounding Grounding language to other modalities Tool Use Teach AI agents to more effectively use existing tools and APIs. Human Preference Alignment Reinforcement Learning from Human Feedback How to scale RL to combinatorially sized language action spaces and messy human preference rewards? Social Value Alignment Aligning AI agents to social commonsense norms and values. Fine-grained Natural Feedback How to build better rewards, i.e. proxy models of human preferences? Human AI Collab Dungeons and Dragons AI Multi-user, situated, collaborative dialogue agents Automated Storytelling Storytelling with an AI Explainable RL Agents RL agents that can explain their actions AI Assisted Content Generation AI collaborators to help humans make (creative) content.
