Research Notes

I write primarily to record and clarify my own thinking. Many notes begin with a question that I explore through reading and discussion, often using AI to test ideas, organize material, and refine the writing. I share them here when I think the resulting ideas or insights may also be useful to others.

01

Beyond Turn-Taking: Toward Models That Listen, Speak, and Act

Full-duplex models, asynchronous reasoning, tools, and the two clocks of interactive intelligence.

02

Training the Assistant: How Post-Training Shapes an LLM’s Character

Post-training as behavioral specification: what SFT and reinforcement learning may shape beyond local response quality.

03

What Is Creativity When AI Can Generate Anything?

What remains scarce when generation becomes cheap: problem choice, distant connections, and judgment.