(deleted)

@khoribe3 That's all very nice, but now what? These feel stuck in their dimension. How can they cross the open ended ladder of complexity into new levels of emergence...if that makes sense.

Automated search in large parameter space to Particle Lenia through ASAL x.com/hardmaru/statu…

@andrewwhite01 Could you point new people to the relevant context of this tweet? I want to know the details.

Google is crushing it lately. We could never make much progress with Gemini about six months ago, but provisioned throughput and the new flash model are so good. Not GA yet, but we should be able to get 100 pages per second for $5k / mo with #3 LLM. Insane progress

Want to understand better the details of Lenia, Flow Lenia and other continuous CAs ? 👾 Take a look at this amazing course by @VassPapas @EPFL_en (with a physics perspective) ➡️frotaur.notion.site/Course-8-Lenia… 📺youtube.com/watch?v=nMoQqp…

@chofski Should visit Belgium to feel better. The only country in the world where there is still no internet on any train. It is one way to avoid the problem.

How is it that the Internet is unusable on public trains in the UK. I remember the good old days of 3G being more responsive! Is it just the web becoming more and more bloated with dependencies?

This is a dead land. Glad most academics, science and tech enthousiasts have moved to 💙. Already more followers than here. network effects.

This place is dead or what?

@kimmonismus @DoctorJosh In bright scenes the robot body feels underlit. Next I'd have to download and slow-mo frame by frame to dissect the interactions with the human. Pretty sure there are artifacts there too

@kimmonismus @DoctorJosh Lack of shadow where the legs touch the ground is super suspicious/unnatural looking.

How did Unitree get so good?

@ellisk_kellis These tweets are so much better than the actual paper abstract. Why is that? Is it on purpose? Article abstracts are not meant as fiction novel appetizers/cliffhangers. I dont know what they teach in scientific writing these days, but I don't like this trend

This suggests a hybrid approach which first does induction to generate candidate programs, and then does transduction if none of the programs work. Our hybrid architecture solves 54.37% of ARC-public-validation and 69.5% of ConceptARC (new SOTAs, we think)

New ARC-AGI paper @arcprize w/ fantastic collaborators @xu3kev @HuLillian39250 @ZennaTavares @evanthebouncy @BasisOrg For few-shot learning: better to construct a symbolic hypothesis/program, or have a neural net do it all, ala in-context learning? cs.cornell.edu/~ellisk/docume…

<< Climate activists shouldn't block the roads. >>

@rfleury It's crazy. Tells a lot about the current state of the industry. Totally moronic topic to go viral.

It is hilarious how controversial this is

Early return is actually not great

@DNAutics I was also surprised how much is still left on the table in what is considered the flagship of an AI assisted IDE. It's all very meh. They have no moat at all imo.

basic feature of an ai code editor is that it should be able to scan subdirectories and make suggestions, not hallucinate what it thinks should be in those subdirectories. #cursorai

@HolgersenTobias @McKayHDavis @Modular General rule: build your career around the cross product what you are “most useful for” and what you love; but balance yourself by making sure you do enough of what you love to stay happy and remain sustainable and scalable.

@mgubrud @tjarnikova @NobelPrize In the life sciences the first author and last author places are precious. They just are. so people are valid in their assumption in so far as to ask the question and expect some explanation. Respectfully, It is your experience outside life-sciences academia that is at odds.

@tjarnikova @craft_links @NobelPrize Again, you cannot assume that the "first author" is the one who deserves the Nobel. In my experience, that would be the last author. No, that is not always the case either, and the conventions may be different in different communities. But you cannot make that assumption.

Congratulations to our 2024 medicine laureate Victor Ambros ✨ This morning he celebrated the news of his prize with his colleague and wife Rosalind Lee, who was also the first author on the 1993 'Cell' paper cited by the Nobel Committee. #NobelPrize

@mgubrud @NobelPrize Usually, the postdoc or student is also the intellectual force. PIs, usually, are very much overrated and more often than not leach on their students. Successful PIs, usually, run a sweatshop. First hand experience in multiple wet-labs. Yes I'm bitter

@NobelPrize OK, folks, so the way this works is that the LAST author is usually the senior scientist and the real intellectual force behind the work. The FIRST author is often a student or postdoc who did most of the actual work. I don't know in this case but that is the usual pattern.

😱 La violence de la gauche encore et encore. Cela fait de nombreuses semaines que la gauche est très nerveuse de voir l’opinion public basculer une nouvelle fois. Ils agissent avec violence et intimidation en n’hésitant pas à instrumentaliser à nouveau le conflit au Proche Orient. 🤯 Hier, le siège du @MR_officiel a, à nouveau, été pris violemment pour cible avec une vitre brisée à coups de marteau, des fumigènes dans le bas du bâtiment (voulaient ils y mettre le feu ?) et des tags plus haineux les uns que les autres. 👎 C’est l’œuvre de soi disants militants anti fascistes et pro palestiniens. En réalité, certains sont juste des petites racailles ignorantes qui cassent pour exister. Elles sont plus fortes pour s’acharner sur une vitre à coups de marteau que de venir débattre arguments contre arguments car ils n’en ont pas. ❌ Que les organisateurs ne se réfugient pas derrière le fait d’une minorité qui aurait infiltré le cortège car nous avons les images du rassemblement durant lequel notre bâtiment est attaqué sous les cris et les chants de l’ensemble du groupe. ‼️ Du personnel était encore à l’interieur ! ⁉️ La gestion administrative de cette manifestation pose également question ! Elle a été tolérée par le Bourgmestre PS de Saint Gilles, ce concept bizarre situé entre l’autorisation et l’interdiction, et qui n’existe pas dans la Loi. Au lieu de bloquer ces manifestants qui devaient être une petite centaine, on a préféré les regarder déambuler et casser. 👉 Je prendrai personnellement toutes les initiatives pour déterminer les responsabilités car je ne peux plus supporter qu’à Bruxelles et en Wallonie ce soient les voyous qui fassent la loi. 🕊️ En ce qui concerne le conflit au Proche Orient, le #MR est la seule formation politique qui a une approche équilibrée reconnaissant le droit d’Israël à se défendre, en particulier contre les organisations terroristes, tout en condamnant les ripostes excessives et en exigeant un cessez-le-feu immédiat. ⚠️ J’insiste sur le fait qu’attiser la violence ici comme s’y prête une certaine gauche ne va pas amener la paix là bas mais au contraire risque de déboucher sur de grandes difficultés chez nous. Qu’attendent ils pour arrêter leur surenchère absurde ? Un drame ? ✋ Nous voulons la paix, ils veulent des votes. Nous aimons les peuples, ils détestent l’occident. Ils sont dans la haine. Nous serons toujours dans la paix et la raison. #PresidenceMR #fierdetreliberal #LAvenirSEclaire

@rikarends That's one hell of a bet? I quit writing Rust. Imo, for 99% of tech businesses Rust is the wrong choice and they should choose a GC language. I'm not convinced there is a big commercial market for targeting Rust devs. But happy to be wrong here.

We have the IDE, we have the UI, we have the AI apis everywhere. I don't have to support anyone elses languages, i can focus 100% on Rust. So yea. Doable goal no?

I really hope i can make AI make writing Rust a bit less painful. If you achieve that, you win the game imho. You get hard typed optimal code AND convenience. Maybe i shall make that my goal.

Since Zig is in the news, let me self-promote my Learing Zig blog series: openmymind.net/learning_zig/

Proud to share that my wife and I have pledged $300,000 to the Zig Software Foundation. Zig has been an absolute joy to use but more importantly I think their culture, vision, and project structure is a bright star for the future of systems programming. mitchellh.com/writing/zig-do…