Benchmarks in Leipzig
36 points by root-parent 49 minutes ago | 11 comments

zerobees 29 minutes ago
I know that people with strong feelings one way or the other will comment here, but note that this is specifically about problems with known answers that can be found in public data (and thus, likely in the training material).

This is an interesting result, but as I understand it, it's not about solving frontier challenges (which LLMs can evidently do too, but that's not what's tested here). It's closer to "can a mathematician (blindly) write exam questions you can't cheat on using an LLM". "Blindly" in the sense that they can't adjust the problem ahead of the time until they get a model to fail.

The conclusion in the paper is: "The concept of writing exercise-style benchmark questions based on publicly accessible research has reached its limits when it comes to the best-performing available models."

reply
lightningspirit 7 minutes ago
I think most of the value LLMs provide comes from connecting the dots between unsolved questions and patterns or structures that have already been demonstrated, which accelerates research.

Now, reasoning in the sense of making truly original discoveries, as Einstein did with the field equations, is a different story for current LLMs.

reply
qsort 24 minutes ago
These are the results from the website they link in the paper:

https://math.sciencebench.ai/benchmarks

I take the "2 unsolved" claim to mean "not solved by any model in any configuration in any stage with any number of attempts", the "benchmark results" are much lower. To be clear: it's extremely impressive, I still remember I was in utter disbelief when models started solving AIME problems, and this is obviously several levels above that.

It's also interesting that OpenAI models perform that much better on math and math-adjacent stuff. I assume this comes down to differences in post-training?

reply
tux3 11 minutes ago
If you're trying to compare what the models are good at, important to note that the different models did not run with the same settings. In one case they also retried with GPT until it answered all the problems but did not retry with the other models.

GPT has 5 effort settings and they picked the highest (xhigh). Claude has 5 and they picked the middle one to avoid having to retry when it timed out. Gemini has medium or high effort and they picked medium.

reply
root-parent 48 minutes ago
"...Between April 1 and May 15, 2026, a group of 49 mathematicians compiled a dataset of research-level mathematics questions with known answers... We present the resulting collection of 100 questions....We evaluated these questions in three stages: a single attempt by five state-of-the-art LLMs....we concluded Stage 3 with only 2 unsolved questions. This demonstrates that the mathematical reasoning capabilities of LLMs are becoming impressive..."
reply
rabidvermin 33 minutes ago
mathematics questions with known answers...

... that are therefore liable to be in the training data?

reply
criemen 26 minutes ago
Partially, 2.2 Submission workflow W2 deals with this:

> Stage W2 The five project-active models, see Table 2, attempted the question. Their answers were compared to the original answer by an LLM judge. If at most three models answered correctly, the contributor could proceed.

So "trivially contained in the training data" is excluded, as then all models could/should easily come up with the solution.

reply
andy99 27 minutes ago
“In the training data” isn’t really relevant for a modern LLM. The better question would be are they solvable using known techniques that have been fine-tuned in.

A simple example, as a non-mathematician: I’d expect a well trained LLM to be able to solve any integral that can be solved with integration by parts. I would be much more interested to see it solve one with no know solution using some novel technique.

Obviously this doesn’t really lend itself to making a benchmark, but if something is solveable by a known technique, and the LLM has has some kind of RL training re using that technique, seeing a solution isn’t too surprising.

reply
fc417fc802 31 minutes ago
I had the same thought, because even if the exact solution doesn't appear there's a notable difference between performing a literature search versus solving something de novo. But I think perhaps this benchmark wasn't meant to exclude the former and that the point may have been to test the ability of the model to accurately interpret and synthesize relevant output for research level mathematical problems at all.
reply
tossandthrow 24 minutes ago
I can recommend reading section 2 of the paper.

The goal was not to define unsolved problems.

But as such, the problems are also not previously published problems.

This seems quite reasonable IMHO.

reply
openclawclub 24 minutes ago
[flagged]
reply
Towaway69 26 minutes ago
As long as it's not conscious, we're safe.
reply