Hacker News: ersiees

New comment by ersiees in "Grok 4.6"

ersiees — Thu, 13 Aug 2026 10:45:04 +0000

Sub two hour marathon is a more recent example of this

New comment by ersiees in "Muse Code and Muse Spark 1.2"

ersiees — Thu, 06 Aug 2026 13:14:27 +0000

Meta, please go back to publishing weights. Your models aren't top-tier, this wouldn't hurt you at all at your current position in the rankings. They wouldn't have the best open-weight models, but the best western open-weight models.

After automation: what if the work of tasking LLMs stays?

ersiees — Mon, 25 May 2026 16:14:29 +0000

Article URL: https://every.to/p/after-automation

Comments URL: https://news.ycombinator.com/item?id=48268534

Points: 1

# Comments: 0

Anthropic is Down

ersiees — Tue, 03 Feb 2026 15:48:03 +0000

Article URL: https://updog.ai/status/anthropic

Comments URL: https://news.ycombinator.com/item?id=46872481

Points: 154

# Comments: 144

New comment by ersiees in "Poison Fountain"

ersiees — Sun, 11 Jan 2026 17:59:22 +0000

Isn’t it too late for that? Won’t that rather cement the oligopoly we have right now?

New comment by ersiees in "Splash-free urinals: Design through physics and differential equations"

ersiees — Mon, 14 Apr 2025 03:49:11 +0000

I think the European urinals already follow the proposed designs closely. They only compare to very old chunky designs.

New comment by ersiees in "Why Fennel?"

ersiees — Sun, 13 Apr 2025 23:34:12 +0000

I don’t love fennel, it usually dominates the whole taste of a dish for me

New comment by ersiees in "How much do you think it costs to make a pair of Nike shoes in Asia?"

ersiees — Wed, 09 Apr 2025 14:54:31 +0000

So, 100$ Nike shoes will soon be 125$.

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Fri, 10 Jan 2025 12:11:45 +0000

Author here!

Thanks :)

1. Only for the first version, not for this version. I am sorry! 2. Yeah ours is guaranteed ok, as we wrote code to generate it basically just from plain torch ops. The code to run inference is available, just not the training code and data generation. 3. We have put it to work on time series data, which is very business relevant for example https://github.com/liam-sbhoo/tabpfn-time-series, and we have a table in the Appendix with all datasets we evaluate on in our main analysis to give you some ideas for possible datasets.

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Fri, 10 Jan 2025 12:07:17 +0000

Yes, there are normalizations applied before the features are fed to the neural network. Additionally, the neural network is trained on a very diverse set of artificial datasets.

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Fri, 10 Jan 2025 12:06:36 +0000

Author here: The new introduction of attention between features did make a big impact compared to the first variant of TabPFN. The old model handled every feature like it was completely different to be feature 5 vs 15, but actually features are typically more-or-less permutation invariant. So the logic is similar to why a CNN is better for images than an MLP.

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Fri, 10 Jan 2025 12:04:22 +0000

To put it very simply, the trick is that while the others train a new model for each problem, TabPFN is pre-trained to handle any kind of problem on the fly.

To draw a parallel to NLP: previously people trained a neural network for each kind of text classification they wanted to do, but then LLMs came around that pre-trained to learn to perform new tasks on the fly. Similarly, TabPFN learns to do new tasks on the fly just from the context (dataset) given.

Training and prediction in these models is by default one and the same, similar to how the prediction of the next token in an LLM is not split into learning from context and then doing the actual prediction. There is a way to split this even up, though, then the predictions, I believe, take something like 1/10s for medium-sized datasets.

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Fri, 10 Jan 2025 08:41:47 +0000

Tuned per dataset

New comment by ersiees in "Show HN: TabPFN v2 – A SOTA foundation model for small tabular data"

ersiees — Thu, 09 Jan 2025 23:16:34 +0000

No, it is *much* stronger, a different architecture and scales to 10x the number of examples. It can also do regression now, and handle categorical features. Please, have a quick look at the abstract before making such claims.

TabPFN-V2

ersiees — Thu, 09 Jan 2025 09:55:44 +0000

Article URL: https://www.nature.com/articles/s41586-024-08328-6

Comments URL: https://news.ycombinator.com/item?id=42643609

Points: 4

# Comments: 0

New comment by ersiees in "Explaining Large Language Models Decisions Using Shapley Values"

ersiees — Sat, 28 Dec 2024 10:28:05 +0000

Another clue: there is no way to download the latex, while you can if someone uploaded the latex on arxiv.

Show HN: Lektor.lol, a free, hacky promptable writing assistant (no sign up)

ersiees — Wed, 11 Dec 2024 16:51:37 +0000

Article URL: https://lektor.lol/

Comments URL: https://news.ycombinator.com/item?id=42389761

Points: 1

# Comments: 0

New comment by ersiees in "Bluesky is currently gaining more than 1M users a day"

ersiees — Sat, 16 Nov 2024 22:47:31 +0000

What is the alternative to AI twitter? Is there some incumbent where it moves?

New comment by ersiees in "Λ-2D: An Exploration of Drawing as Programming Language"

ersiees — Fri, 08 Nov 2024 17:12:47 +0000

And there is scratch right?

New comment by ersiees in "I couldn't find a free, no-login, no-AI checklist app–so I built one"

ersiees — Sun, 03 Nov 2024 17:54:03 +0000

Why is it only for 7 days available?