Sakana AI's MASS method has a single language model solve tasks, judge the results and refine its workflows, reaching a win rate close to 70 percent against the base model after two rounds. The catch: when the model misses the same error in all three roles, it can treat that error as progress and reinforce it in training.
Microsoft researchers found autonomous AI agents get stuck in 'idea basins', local solution regions created by their growing chat context. Their Fork-and-Flush method wipes the chat history but keeps the workspace, yielding up to 66% better results on the same compute budget.
Microsoft Research's StudentSim models how individual learners might engage with AI tutors, so the tools can be evaluated and improved before reaching real students. It provides feedback on tutor performance in subjects such as chess, writing and mathematics.
Alibaba's Qwen team unveiled QwenGyre, an online reinforcement learning framework that trains coding agents through unmodified, black-box harnesses. It targets rollouts lasting hours and processing close to a million tokens.
The latest Import AI newsletter covers creativity and recursive self-improvement, open-weight models said to trail Chinese ones, and AI in space.
A study presented in early October suggests that about ten minutes of AI use is enough to reduce people's persistence when facing a problem. Long-term effects of AI on intellectual abilities remain poorly understood.