Hunter McNichols

Identifying key traits in human-AI collaboration could redefine how we understand and teach the skills necessary for successful teamwork with AI agents.

LLMs can now generate targeted, improved versions of student writing that are both rubric-consistent and closely resemble the original, thanks to a novel noise-steering method.

LLM-generated personalized math problems often fail on realism and authenticity, but can be substantially improved with a multi-agent validation and refinement loop.