全部动态

https://x.com/i/article/2118746079923437568

https://x.com/i/article/2118746079923437568

New on Open Math OS: Mistral Large 4 by @MistralAI. 1.05T open weights. 1M context window. Available now to all approved teams.

Frank Calegari has a great post on Terry's blog: https://terrytao.wordpress.com/2026/11/01/mathematics-has-changed-lets-act-like-it/ I think it aligns most closely to what I've been thinking and trying to say (and I believe communicates it better than I've been able to). Some highlights: "I hope AI can help mathematicians break out of the straitjacket of being constrained by the first natural conjecture they cannot prove, and work on unraveling the deeper mysteries woven throughout our subject." "The natural grooves of our profession have very much been bent toward more narrow technical expertise than a bigger vision. AI allows us an amazing opportunity to broaden ourselves as mathematicians." "There are so many problems in mathematics that are still so fresh and vital but were previously untouchable that the opportunity to approach them is massively exciting. We need to be more ambitious, to create bold research plans that weren’t even conceivable a year ago." "The difference in reaction between the optimists and the pessimists is that, for the pessimists, the game is over. For the optimists, the game has just changed." "[T]here now exists an amazing new tool to study mathematics. To turn away and say that the community of mathematics as it currently exists is more important than mathematics itself would be a terrible mistake." "[I]t takes one look at the arXiv to see many people in our profession rushing out AI-assisted papers in a somewhat unsightly rush to be the “first” to prove things." [Aside: mere days after the qRH result, I had something like a half-dozen emails in my inbox, some from amateurs but some also from tenured faculty, with Lean formalizations of nonvanishing on Re(s)> 7/8-1/10^k for various large values of k. What the hell are you even doing? That kind of thing may have been valuable years ago, when it was difficult and signaled deep understanding and great technique. Today? Who the hell cares?] "To me [this is back to Frank now], someone who chooses a worthwhile problem, obtains a proof completely using AI, understands it thoroughly, and uses it to develop interesting further questions or directions has made a substantial mathematical contribution." "If you, as a senior mathematician, are stressed about being scooped by AI, then you are not being ambitious enough." [On hiring:] "I am looking for a narrative of where you want mathematics to go, the technical ability to explain why this may be possible, being able to converse about many areas of mathematics, and a demonstrable record of having done all of this in the past." "What I want a graduate student to achieve at the end of their PhD is a view of mathematics that extends well beyond their discipline, and yet with demonstrable technical skills on at least some class of problems." "I advocated above that we need to evaluate work as if AI was involved. But this does not imply that we don’t need to teach students to do deep mathematics without AI just as the existence of computers doesn’t mean we should no longer teach children arithmetic." [I've often said similar things as well...] "All of my friends in the humanities truly believe that having mathematicians in society both as custodians of one of humanity’s oldest intellectual activities and explorers of the frontiers of knowledge is important. I understand that some part of the culture we so deeply care about will disappear, and as humans, an emotional response to that is very natural. But this is a time to explain to everyone that there are so many great new opportunities, possibilities, and reasons to be optimistic. We are on the verge of a new era of mathematics, and I am excited to see what happens." Bravo Frank!

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: https://rikiyat.github.io/blog/weight-space-programming/

We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math. It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training. We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! 🧵 Blog: https://rikiyat.github.io/blog/weight-space-programming/