Vaclav Kosar

@ddebowczyk Ad composable programs: But what can you do with DSPy language that cannot be done with normal Python?

The most powerful capability of DSPy is composable language programs. Prompt optimization is great, but modules are the essence of DSPy value for me. Also, I think TextGrad approach to optimization might be a better long term approach (but I'm not sure until I reimplement both and compare).

I've been experimenting with DSPy like modules in Instructor and atm I'm not clear how to proceed. So far I designed modules from scratch, but I'm concerned over all the novelty introduced by Instructor. So maybe modules could be based on some existing approach, familiar to Laravel or Symfony devs, like Laravel actions, jobs, or Symfony workflows (but they are all framework specific). Is there any simple (!) PHP package that allows defining data processing flows in a convenient way (+handling failures)? Any ideas X?

The Czech Republic has chosen UK firm @RollsRoyce SMR after assessing seven potential technology suppliers for its proposed small modular reactor programme #nuclear tinyurl.com/yc63hv23

Today, we release several Moshi artifacts: a long technical report with all the details behind our model, weights for Moshi and its Mimi codec, along with streaming inference code in Pytorch, Rust and MLX. More details below 🧵 ⬇️ Paper: kyutai.org/Moshi.pdf Repo: github.com/kyutai-labs/mo… HuggingFace: huggingface.co/kmhf

@ddebowczyk Here is work around for this DSPy bug - Wrap into exception removing predictor: #issuecomment-2357670013" target="_blank" rel="nofollow noopener">github.com/stanfordnlp/ds…

@ddebowczyk Ad structured outputs: I am testing TypedPredictor and DSPy has this bug: If the LLM fails to adhere to the schema, the entire evaluation or optimization process stops, which is a major problem. I have to work around this. This is a strange thing to have.

@cognitivecompai If self-sustaining then it would save us. Self-sustaining part is hard, but solvable. Also second planet is would reduce zero-sum thinking. A new frontier would allow for new societies like during discovery of America. Space mining could also increase rare resources.

This is true - but I don't think a settlement on Mars will save us.

@staysaasy You should test this. From time I will listen. Even though, I was distracted by podcasts recently and I will be reducing listening time a bit.

Would you listen to a StaySaaSy podcast?

Thiel at All-In Summit: "we probably tried to do too much investing in Europe over the years it's always sort of a junk it sort of it's a nice place to go on vacation as an investor...the US ...where people do new things"

I’ve just finished reading the Draghi Report, and to my surprise, 4 out of the 5 policies I proposed in my EU/ACC open letter were included almost word-for-word. It's encouraging to see this alignment, but there’s still work to be done. Signing the open letter is a way to keep the pressure on decision-makers and ensure that these crucial ideas continue to gain momentum. Let’s make sure our push for progress doesn’t stop here. 📄 Read and sign here: docs.google.com/document/d/1jF… My latest article: x.com/Duarteosrm/sta… #EUACC #Draghi #eu/acc

Terence Tao's take on o1. TT is a mathematician won the Fields Medal in 2006 and is widely regarded as one of the greatest living mathematicians.

"In summary, 01 represents a paradigm shift from memorize the answers to memorize the reasoning", but still "fitting a curve to a distribution" So, no gleam of AGI yet.

Thanks 🫶to all our sponsors for a more than generous amount of cloud GPU credits! (European mind can’t comprehend… 😉).

Losing mindset: * Setting (lower) expectations * Talking about how things could be worse paired with inaction Winning mindset * Setting (high) expectations and figuring out how to deliver * Talking about how things could be better paired with action to get there.

OpenAI o1 — our first model trained with reinforcement learning to think hard about problems before answering. Extremely proud of the team! This is a new paradigm with vast opportunity. This is evident quantitatively (eg reasoning metrics are already a step function improved) and qualitatively (eg faithful chains of thought make models interpretable by letting you “read the model’s mind” in plain English). One way to think about this is that our models do System I thinking, while chains of thought unlock System II thinking. People have discovered a while ago that prompting the model to “think step by step” boosts performance. But training the model to do this, end to end with trial and error, is far more reliable and — as we’ve seem with games like Go or Dota — can generate extremely impressive results. It’s still early days for the o1 technology. It provides new safety opportunities which we are exploring actively, including on reliability, hallucinations, and robustness to adversarial attackers. For example, we’ve seen great uplift in our safety metrics by letting the model reason about policies via chain of thought. Its accuracy also has huge room for further improvement— for example, from our launch post, our model achieved 49th percentile / 213 points in this year’s competitive programming Olympiad (IOI) under human conditions of 50 submissions per problem. But with 10,000 submissions per problem, the model achieved a score of 362.14 — above the gold medal threshold. So the model is capable of even greater outputs than it appears at first glance.

Rowing, apart from cardio, seems to be indeed good for the back, too. Still rowing tiny 2 min sessions every day. Row! Row! 🚣

@maxazoury @rohanpaul_ai Yes, LLMs must hallucinate intuitively. But proving it and writing it up is still a job that they did. The paper seems coherent at a glance. I like the distinction: "hallucination=plausible but incorrect information", because anything novel has some similarity to hallucination

"LLMs Will Always Hallucinate, and We Need to Live With This" - 🤔🤔 Key points from the paper. 👇 🧠 Hallucinations in LLMs not just mistakes, but inherent property. Arise from undecidable problems in training and usage process. Can't be fully eliminated through architectural improvements or data cleaning. 🔬 They use computational theory and Gödel's incompleteness theorems to explain hallucinations. Argue that LLM structure inherently leads to some inputs causing model to generate false or nonsensical information. 🚫 Complete elimination of hallucinations impossible due to undecidable problems in LLM foundations. No amount of tweaks or fact-checking can fully solve this issue. Fundamental limitation of current LLM approach. ------ 🧮 Gödel's incompleteness theorems: 👉 First theorem: Any consistent formal system powerful enough to encode arithmetic contains statements that are true but unprovable within the system. 👉 Second theorem: Such a system cannot prove its own consistency.

🇪🇺 eu/acc A few weeks ago Mario Draghi asked my recommendations for his report that came out today about European competitiveness I had a call with him and summarized my problems with doing business in the EU I wrote this which is included in the report presented to the European Union today: 1. Minimum revenue cut offs for current and new regulation Exempt small businesses with annual revenues below €10 million from complex regulations like VATMOSS, GDPR, the EU AI Act, and certain labor laws. This approach encourages innovation and growth by allowing startups to focus on product development and market validation without the heavy burden of regulatory compliance. Once these businesses surpass €10 million, they will have the resources to comply with regulations, ensuring that growth is not stifled. 2. Simplify starting a pan-EU business with an EU-wide Incorporation (Inc.) business form Currently, starting and operating a business across the EU is complex due to 27 member states, each with its own company registration requirements. To streamline this process and make it easier for entrepreneurs to operate across Europe, there should be a single, standardized business entity that applies uniformly across all EU countries. I call this the European Inc. 3. Start an EU business fully online, no physical offices, notaries, lawyers etc To continue, right now starting a business in most EU member states it’s complicated, very time and resource intensive, and often involves lawyers and notaries. Instead, it should be as simple as going online to a centralized EU website, where entrepreneurs can register their business and details in just a few clicks. The entire process should be streamlined and efficient, allowing businesses to start operating immediately. The EU government taxes and bookkeeping of this business should also be fully online in an EU portal/dashboard. 4. 0% corporate tax for first 3 years of any new business Countries like Singapore have successfully attracted new businesses from around the world by giving them a massive tax discount during the first 3 years of business. Because they know that’s the most difficult time of a business: figuring out what product it makes and if there’s a market for it. That takes pressure off startups and business founders that they can focus on creating a great product and innovating. 5. Change tax on stock options: don't tax when a stock option is exercised, but tax it when the stock is sold The current tax policy in the EU taxes stock options at the time they are exercised, creating a significant financial burden on employees who have not yet realized any tangible financial gain. This approach stifles innovation, discourages entrepreneurship, and places the EU at a competitive disadvantage compared to other regions like the United States. I propose a simple change: Tax stock options when the stock is sold, not when the option is exercised. 6. Don’t see tech or AI as an enemy, but as a burgeoning and essential industry The most popular companies in tech are focused on AI right now for a reason. It’s the next frontier of computing. The European Union seems to consider AI the enemy. Any technology can be used for good or bad. By regulating it even before Europe has made much contributions (Europe has almost no tech companies leading in AI), it has stifled any potential innovation in AI from the start. Apart from the regulation itself, the optics of it make the EU look bad on a global scale. Why would tech founders move to Europe to start a business if the EU is actively positioning itself as Anti-AI? AI has gigantic potential to be used for good: think of the medical field for diagnosis of diseases, generally in programming (it helps programmers to create software faster/better), etc. This goes further than AI. The same applies to tech in general. It seems the EU is on a crusade against technology while not being able to compete in it itself. It feels a case of sour grapes: if we can’t build great technology in EU, nobody is allowed to do so! 7. Teach tech/coding/AI topics in all schools and unis It would help a lot if the EU has a focus on teaching AI and tech in schools and universities. Making the new generation competitive in this field instead. To secure the future prosperity of the European Union, we must prioritize education in technology, coding, and AI across all levels of schooling, from primary education to universities. This strategic focus is not just an educational reform—it’s a critical investment in the future competitiveness, innovation, and economic resilience of the EU.

Growth in GDP, technology, and reduction of red tape prevent societies from weakening and crystallizing into feudal hierarchies without upward mobility. In the long term, through good economics and innovation, EU members can not just make their land flourish but even uncover a new land on Mars's new free frontier. Therefore, it is particularly important to power up the growth.

Yes, it makes a difference We, Europeans, working together to make noise 📣 To spread our voice Yes key people follow us Yes, at 🇪🇺 institutions Yes they want us to be stronger & more visible so we together make Europe a welcoming place for creators, innovators. Amplify us 🇪🇺

@XiXiDu I would bet on the Mars colony before the singularity. AI will have a large impact from now on, but we may not be not near singularity. Is there a scenario you think is realistic? Also the plan encourages forward action. Is singularity scenario more sit and wait?

This is awesome and I fully support it. But I wager to say this is slightly inconsistent with his (and my own) AI timelines. Whether a self-sustaining Mars colony can be built before the singularity happens is questionable. But definitely worth trying anyway.