1. Two things can suck independently at the same time. One thing sucking is not a justification for another thing sucking.

2. Google doesn't pretend that its algorithm has certain capabilities while serving the user a nerfed version

I think (2) is the important one. Anthropic provides benchmarks with their model releases. If they are making changes that make models perform worse on advertised benchmarks, how can this be legal?

It's a good point. I mean I have no idea if they are actually breaking the law, would assume their tc's would dissolve them from any legal jeopardy but it's still not right.

Is this where independent benchmark companies like artificial analysis could help ?

In fact, people accepting that the new thing can suck is why newer things suck more. That's how the frog boils.

Google gives a lot of information to its customers, the advertisers, when it changes things that impact them a lot. Not enough for all of them to be happy, but it does private and public communications on big changes.

Atlantic and New Yorker my two favorite publications for long form articles. Although the Atlantic has got a bit too political for my tastes

At a 3 minute shot, I’d rather use the same time to do a pour over

How long does it take for water to boil? This is slower but still not by much.

If you have a thermoblock espresso machine, the brewing part can be over and done with in all of 1 minute.

But either way, the time consuming parts are preparation and cleanup.

You might as well play "who can memorize the most openings and lines"

Seems dismissive at first, but I interviewed a chess team captain once and he told me they prepared for upcoming matches by learning new lines they wanted to play and studying the lines the opposing school was deemed likely to play.

It is a memory game yes! But still you can choose what to memorise.

It can get complicated quickly if you're actually using it in a production system. At my prev enterprise saas company we had feature flags that could be turned on per customer / per environment (dev, staging, prod) with permission + logging model such that our support team could also toggle flags with history of who turned on what. We also had "per user" feature flags for certain test users at companies and had DSL rules to evaluate the features

this is essentially how bindcraft works for drug discovery: https://www.nature.com/articles/s41586-025-09429-6 (minus the accumulation step)

feedback from af2 folding confidence + structural scoring

Fish sauce is delicious but had to stop using it since it's high in histamine (gives me a stuffy nose) and potentially carcinogenic due to its high levels of nitrosamines

industrialized overfitting is basically what ML researchers do

should go a bit earlier with word2vec, NMT, seq2seq, attention, self attention

Little late to this thread but from my list:

LLM (foundational papers)

* Attention is all you need - transformers + self attention

* BERT - first masked LM using transformers + self attention

* GPT3 - big LLM decoder (Basis of gpt4 and most LLM)

* Instruct GPT or TKInstruct (instruction tuning enables improved zero shot learning)

* Chain of Thought (improve performance via prompting)

some other papers which are become trendy depending on your interest

* RLHF - RL using human feedback

* Lora - make models smaller

* MoE - kind of ensembling

* self instruct - self label data

* constitutional ai - self alignment

* tree of thought - like CoT but a tree

* FastAttention,Longformer - optimized attention mechanisms

* React - agents