Most benchmarks can be trained for as well, so they are over-representative of model's engineering skills.
The entire nature of a benchmark is collapsing some qualitative work (software engineering task, architecture choice, code quality) into a quantitative score which can be optimized for.
Let's say you to render conditionally when `x` is present. You do `v-if="x"`. Now you want to refer to `x.y.z` in the body, while `y` is optional.
Can you be sure that the value is present? How do you check for it? What does Vue need to do to enable this functionality with static type checking with TypeScript?
With `{x && <>{x.y.z}</>}`, it's almost vanilla TS, it's the same as `x && x.y.z` in normal TS. Type narrowings that work in TS are guaranteed to work in React, without the framework having to do anything.
Less framework/library code required to make the same thing work. To me, that's the sign of a better abstraction and implementation.
JSX is a very thin layer of templating logic inside JS. Meaning you have all the language features available while templating. Some arbitrary JS can result in templating.
Vue's DSL is whatever language the developer implemented. Which is probably not enough, depends how much effort they put into it and how good they are in language design. Given that they cargo cult HTML tags to organize components in a pseudo-familiar but not-valid-HTML way, I don't have much confidence in their language design skills.
Scaling law is a power law , requiring orders of magnitude more compute and data for better accuracy from pre-training. Most companies have maxed it out.
Next stop is inference scaling with longer context window and longer reasoning. But instead of it being a one-off training cost, it becomes a running cost.
In essence we are chasing ever smaller gains in exchange for exponentially increasing costs. This energy will run out. There needs to be something completely different than LLMs for meaningful further progress.
Math and coding competition problems are easier to train because of strict rules and cheap verification.
But once you go beyond that to less defined things such as code quality, where even humans have hard time putting down concrete axioms, they start to hallucinate more and become less useful.
We are missing the value function that allowed AlphaGo to go from mid range player trained on human moves to superhuman by playing itself.
As we have only made progress on unsupervised learning, and RL is constrained as above, I don't see this getting better.
In simpler terms - they create an MCP server, essentially an API that the coding agent can call, that can fill in context about previous decisions done by the coding agent earlier in development. Agent equivalent of asking someone who's been working there longer "why is this this way".
This means that the agent will can have context of previous decisions, something that they currently struggle with as they are always starting from a blank slate.
Treating T2D and preventing T2D are completely different things from a dietary perspective. Same way you wouldn't give chemotherapy to a healthy person to prevent cancer
Red meat (a known carcinogen) at the top is gold. All that saturated fat the energy will come from (not from protein or veggies) will probably cause heart problems and plaque formation in arteries, not to mention insulin resistance just from increased FFAs in blood.
Vegetarians and vegans have lower T2D incidence on average FWIW.
i would interpret physical fitness as cardio exercise routine and depleted muscle glycogen stores:
so breakfast is very welcome and without it is not possible to keep up exercise routine
Yes, but it is not syntax. It's a contract with the library. React is completely usable using vanilla JS syntax. Same cannot be said for Vue and Angular.
It feels a bit like talking about apples and oranges in this thread.