rfurmani

I'm one of those two, as mentioned later in the post! As it stands, by mathematical standards, it would be inappropriate for us to be authors on the paper since our role was more like a highly interested referee, but we do take mathematical accountability for it. This all happened very quickly, but we will be sure to continue to polish the paper and make it ready for publication.

Dollars of compute at market rate is what I'd like to see, to check whether calling this tool would cost $100 or $100,000

Sounds like it did not:

> This year, our advanced Gemini model operated end-to-end in natural language, producing rigorous mathematical proofs directly from the official problem descriptions – all within the 4.5-hour competition time limit

Only for that particular board, in general it will be very complex and depend on the shape of the colored regions

I've had a couple bad experiences with Lyft recently, including one time the driver must have clicked that they picked me up while a block away, because I could see the lyft driving to the destination without me. I tried to get a refund since I was obviously waiting my start location the whole time, but the system claimed the drive went from start to finish (even though I wasn't in the car), so no refund.

Completely agree on both counts! I loved those two games and felt Conquests of the Longbow didn't get the recognition it deserves.

On the second point, when I read his book (https://kensbook.com/) I was disappointed to not hear about the magic of the games themselves and the creative process behind them. It became clear that his primary goal was to grow a business, he thought being a game distributor was more exciting, but then was disrupted by Steam, shareware, and online distribution.

I'm building such tools at https://sugaku.net, right now there's chatting with a paper and browsing similar papers. Generally arXiv and other repositories want you to link to them and not embed their papers, which makes it hard to build inline reading tools, but it's on my roadmap to support that for uploaded papers. Would love to hear if you have some feature requests there

I've wanted to but wasn't sure how to keep track of individuals. What works for you? IP Addresses, cookies, something else?

After I opened up https://sugaku.net to be usable without login, it was astounding how quickly the crawlers started. I'd like the site to be accessible to all, but I've had to restrict most of the dynamic features to logged in users, restrict robots.txt, use cloudflare to block AI crawlers and bad bots, and I'm still getting ~1M automated requests per day (compared to ~1K organic), so I think I'll need to restrict the site to logged in users soon.

Very cool! This is also one of my beliefs in building tools for research, that if you can solve the problem of predicting and ranking the top references for a given idea, then you've learned to understand a lot about problem solving and decomposing problems into their ingredients. I've been pleasantly surprised by how well LLMs can rank relevance, compared to supervised training of a relevancy score. I'll read the linked paper (shameless plug, here it is on my research tools site: https://sugaku.net/oa/W4401043313/)

That's fair, though there's not much to say since I'm building it out myself as a benefit corporation. I also have strict opt-out for any communications and a proper privacy policy.

I've also tried to keep as much as I can accessible without login, but I want to protect some of the more expensive features from being spammed.

Without signup up you can:

* explore works (but not chat with them) https://sugaku.net/oa/W4206400500/

* explore authors https://sugaku.net/oa/A5059543195/

* see and share AI answers (eg https://sugaku.net/qna/4e59662a-a938-404e-8c0b-b9dc79e37c29/ and https://sugaku.net/qna/517930ff-42ad-47c5-9d9c-e807d06a8453/)

* prompt for new paper ideas https://sugaku.net/current/papergen/

* see and share these ideas https://sugaku.net/current/papergen/idea/719aed36-8dcd-4fd1-...

The flipside of that is seeing hints of a result that would be really helpful. I still remember how excited I was to stumble on a book from 1931 (The Taylor Series by Dienes) since it had the only english-language proofs of some results by Szego and Polya that I felt could unblock my research. My hope is that this discovery problem can be largely solved.

This is also why I'm not as excited by the focus on pure reasoning and olympiad problem solving in the math and AI space. It's like the early career phase of trying to solve Collatz and Riemann but just repeating work from decades ago.

I'm serving AI models on Lambda Labs and after some trial and error I found having a single vllm server along with caddy, behind cloudflare dns, to work really well and really easy to set up

vllm serve ${MODEL_REPO} --dtype auto --api-key $HF_TOKEN --guided-decoding-backend outlines --disable-fastapi-docs &

sudo caddy reverse-proxy --from ${SUBDOMAIN}.sugaku.net --to localhost:8000 &

As a former mathematician, I found research to be a very winding path. While that can be fun, I felt there's a lot of opportunity to train LLMs and ML models on the corpus of math papers, to try to make research more deliberate and less reliant on talking to the right person at the right time.

This is very much a work in progress but so far you can:

* Browse through similar papers

* Get recommendations for new papers and collaborators

* Chat with papers and ask questions to all the major reasoning models

* Have it come up with future paper ideas (along with references) giving a potential title or collaborators.

My focus very much is on the exploratory stages since that's where a lot of the time is spent, but I intend to integrate more tools for problem solving, writing, and computation.

Wow, Gemini 2 has a 9 step process for the first question https://sugaku.net/qna/1d922e55-fd28-4c70-b3e9-d70c3a05ce1f/

> Here's how you can get exactly 6 liters of water using a 12-liter jug and a 6-liter jug:

1. Fill the 6-liter jug completely.

2. Pour the water from the 6-liter jug into the 12-liter jug. You now have 6 liters in the 12-liter jug.

3. Fill the 6-liter jug completely again.

4. Pour water from the 6-liter jug into the 12-liter jug until the 12-liter jug is full. Since the 12-liter jug already had 6 liters, you will pour 6 liters. This leaves you with 0 liters in the 6 liter jug.

5. Now empty the 12 liter jug.

6. Pour the water from the 6-liter jug into the 12-liter jug.

7. Fill the 6-liter jug completely again.

8. Pour the water from the 6-liter jug into the 12-liter jug until the 12-liter jug is full. Since the 12-liter jug already had 0 liters, you will pour 6 liters. This leaves you with 0 liters in the 6 liter jug.

9. There are 6 liters left in the 6 liter jug.

As for Rs in strawberry, trying a bunch of models side by side only Sky T-1, Gemini 2 Flash got it wrong! https://sugaku.net/qna/792ac8cc-9a41-4adc-a98f-c5b2e8d89f9b/

Simple questions like 1+1 can also be fun since R1 goes overboard (as do some other models when you include a system prompt asking it to think) https://sugaku.net/qna/a1b970c0-de9f-4e62-9e03-f62c5280a311/

And if that fails you can ask for the zeros of the ζ function! https://sugaku.net/qna/c64d6db9-5547-4213-acb2-53d10ed95227/

Absolutely agree. There's some interesting articles in a recent [AMS Bulletin](https://www.ams.org/journals/bull/2024-61-02/home.html?activ...) giving perspectives on this question: what does it do to math if there's a strong theorem prover out there, in what ways can AI help mathematicians, what is math exactly?

I find that a lot of AI+Math work is focused on the end game where you have a clear problem to solve, rather than the early exploratory work where most of the time is spent. The challenge is in making the right connections and analogies, discovering hidden useful results, asking the right questions, translating between fields.

I'm getting ready to launch [Sugaku](https://sugaku.net), where I'm trying to build tools for the above, based on processing the published math literature and training models on it. The kind of search of MR that you mentioned doing is exactly what a computer should do instead. I can create an account for you and would love some feedback.

You're definitely not the only one! We [1] have been in-person in San Francisco predominantly, at first due to visa sponsorship rules, but also due to the energy you get at a fast moving early stage startup, allowing us to scale super fast. We've had to constantly evaluate whether we are making the right decision, especially as we say no to really talented people who are remote, but some of our early engineers put their foot down and let us know that they chose us /because/ we were in person. And time and again we've seen that there's plenty of people like you who want to be around people and feel connected to everything happening, especially those who are looking at series A/B startups. Frankly, it also just seems that you need to be a lot more rigid, focusing on specs, structure, documentation, when remote-first and that is not as fun when hacking. We of course are flexible about work from home, or traveling to see family or be in other places, but last time I was with family I definitely noticed that working over zoom and slack was way more exhausting.

[1] https://parafin.com/ https://www.linkedin.com/company/buildparafin

It's not quite so simple, since we are looking at single examples. Alexander was predated by leaders (Cyrus the Great) who ruled in a much more progressive gentle way. Recent history has shown major powers performing very cruelly (genocides under colonialism) and there's definitely at least minor world leaders who would want to be just as cruel. So the question becomes more of why was it effective and strategic to be cruel at that scale in the past but not in the present world, and I would attribute that to technology: better communication, travel, and interconnectedness of trade.

Parafin | San Francisco | ONSITE, VISA | https://parafin.com

As featured in yesterday's Wall Street Journal: https://www.wsj.com/articles/former-robinhood-employees-laun...

Small businesses are the backbone of our economy, and we want to enable the very platforms they transact on such as on-demand marketplaces and vertical SaaS offer up a full-suite of financial services to them.

Our first product that is live today offers capital-as-a-service for online platforms. We currently power the capital programs of platforms ranging from Series B to post IPO companies. We are a tight knit team of engineers and designers, coming from Robinhood+Affirm+CapitalOne+Uber+CERN, excited both about helping hundreds of thousands of small businesses grow and thrive, and iterating on products that leaders at top companies have started to use every day.

You can see our roles at https://www.parafin.com/careers or reach out at info@parafin.com.

Our tech stack is AWS + Scala + React.

Excited to finally announce and answer question! Also featured in the Wall Street Journal today: https://www.wsj.com/articles/former-robinhood-employees-laun...

Co-founder here. It's been an incredible time to build fintech infrastructure to address the most burning needs of small businesses. We've been able to launch with some incredible partners and have our update at: https://www.parafin.com/blog/announcing-parafin

Parafin | San Francisco | React Engineer, Infra Engineer | https://parafin.com

Small businesses are the backbone of our economy, and we want to enable the very platforms they transact on such as on-demand marketplaces and vertical SaaS offer up a full-suite of financial services to them.

Our first product that is live today offers capital-as-a-service for online platforms. We currently power the capital programs of platforms ranging from Series B to post IPO companies. We are a tight knit team of engineers and designers, coming from Robinhood+Affirm+CapitalOne+Uber+CERN, excited both about helping hundreds of thousands of small businesses grow and thrive, and iterating on products that leaders at top companies have started to use every day.

You can see our roles at https://www.parafin.com/careers or reach out at info@parafin.com. Our target roles are below, but we are biting off a big space and are eager to talk to anyone who is excited to add value!

* Product Engineer (Web)

* Infrastructure Engineer (AWS/CDK)

* Backend Engineer (Scala)

* Business Operations Specialist

* Customer Experience

* Account Manager

* Office Manager

* Executive Assistant

It must be the command-line calculator bc

It’s not just about funding your account, it’s every single trade. When you sell a stock and then immediately buy another one, the cash from your sell doesn’t actually land for a few days.

That said, I do wish I were at AIM's meeting on the Riemann Hypothesis :-) https://aimath.org/rh2018/

The lecture videos are online and are a good way to get both historical context and connection to how people are thinking about it currently.

This seminar talk on how to fail to prove RH was also great https://www.math.rutgers.edu/news-events/list-all-events/ica...

There's been a lot of research that has assumed it to be true and built on top of it to find deeper implications.

Still, a number of people who've worked most closely with the Riemann zeta function, even the ones who led the computations and numerical verifications, have expressed doubts on whether the Riemann Hypothesis is actually true: https://arxiv.org/abs/math/0311162

I love this quote:

> A meeting on how to extend the GPY method was immediately organized at the American Institute of Mathematics in San Jose, California. As a bright-eyed and bushy-tailed grad student, I felt extraordinarily lucky to be there among the world’s top experts. By the end of the week, the experts agreed that it was basically impossible to improve the GPY method to get bounded prime gaps. Fortunately, Yitang Zhang did not attend this meeting. Almost a decade later, after years of incredibly hard work in relative isolation, he found a way around the impasse and proved the experts wrong. I guess the moral of my story is that when people organize meetings on how not to solve the Riemann hypothesis (as they do from time to time), don’t go!

I've heard very good things about Visual Complex Analysis, as well.

This Numberphile video on the Riemann Hypothesis is also very good: https://www.youtube.com/watch?v=d6c6uIyieoo