Interesting. My intuition here is quite the opposite, that math is vastly overpowered for physics. I don't think we invest a lot of work in figuring out math that isn't useful to describe the natural world.

Recommendation: rig up the mcp and have claude or codex trawl the corpus for the kinds of ideas you might care about.

``` claude mcp add --scope user lacuna-research -- uvx lacuna-research-mcp ```

or

``` codex mcp add lacuna-research -- uvx lacuna-research-mcp ```

One part of the explanation is that we focus primarily on AI research, and a substantial chunk of physicists gravitate (heh) towards AI research, so we end up capturing more of their work.

Another part of the explanation could be that (AI +) physics papers are just easier to access, so we end up processing more of those.

I came up with that name, and fair eyebrow. Think of an "atom" as a unit of a working vocabulary (a "token"), not a Platonic form. It's a way of compressing the corpus (paraphrases of the same idea across papers get merged into one symbol, a bit like a tokenizer's vocabulary).

We also talk about some limitations of this framing in page 11 (towards the top).

The one thing I could recommend however is rigging up the mcp (https://github.com/tiptreesystems/lacuna-research-mcp) to your code agent and trawling the database for ideas worth thinking about.

> I find that often a big cluster of papers around a particular topic, none of which seems to move the needle much, is actually an indication that we need a new approach and writing more and more incremental papers just digs deeper into a local minimum.

Yes, I think this can work. We use a similar idea to categorize what papers have been influential or foundational -- look at the ones that have changed the trajectory (in time) of a research direction significantly.