Source archive
AI systems, agents, models, and scientific research from Microsoft Research. This source page gives readers and crawlers a stable route for the latest Microsoft Research coverage.
Indexed briefings
6
Latest source-linked updates, ordered newest first.
Latest
Training AI agents with reinforcement learning can be challenging because their tools, context, and decision-making are managed by complex frameworks.
Orchard is an open-source framework for the research community to train and evaluate AI agents across task types.
Computer-use AI agents struggle with multi-step workflows like email and customer support.
Short chart specifications are easy to write, but often produce uninspiring results.
AI agents often fail because their instructions, or skills, are manually modified with no guarantee of improvement.
AI agents can't remember past conversations.