Manling Li

Manling Li is an Assistant Professor of Computer Science at Northwestern University. Their research focuses on multimodal learning, vision-language models, and embodied agents. They explore challenges in spatial reasoning, knowledge hallucination in LLMs, and event understanding. Their work often bridges AI, computer vision, and natural language processing to advance autonomous agent capabilities and video-language understanding.

Research interests prominently feature multimodal benchmarking (e.g., EmbodiedBench), video-language analysis (Hourvideo), and mental health applications via self-play systems like MentalArena. They investigate foundational issues like knowledge overshadowing in LLMs and develop techniques for 3D layout optimization (LayoutVLM).

No scientific awards are explicitly listed in the provided information.

No listed advisees or grants are currently available. Their contributions emphasize cross-disciplinary collaboration between computational models and real-world tasks, such as assembly instruction grounding and long-form video comprehension.

Labs or team affiliations are not explicitly detailed in the provided information, though their work suggests involvement in AI/ML research groups focused on embodied intelligence and multimedia systems.