Royal Road

Why I made it

Machine-learning research is a tangle of thousands of papers, and it is hard to see the thread: which ideas a method rests on, and which ones grew out of it. I wanted a map of that lineage — a road from the foundations to the frontier of any topic — paired with honest reviews so you can judge each step for yourself. I built it to learn the field this way, and I'm sharing it in case it helps you do the same.

How to use it

Open the explorer and pick a destination — say, Flow Matching or Transformers. Royal Road draws the road to it: the foundational papers on the left, the frontier on the right, in the order they were built. Click any paper to read its review, or use “Trace the road to this paper” to redraw the map around it. Prefer to browse? The index of papers is a searchable list of everything reviewed so far.