Hacker Newsnew | past | comments | ask | show | jobs | submit | heltale's commentslogin

Has anyone tried Migadu? I was wondering whether it’d be a good option for if I wanted to use it for my business.


Same! I’m trying to find small models that can embed effectively to enable BM25/hybrid search over a large number of documents for a personal information repository. Ideally, it should run on consumer hardware.

bge-small-en-v1.5 is one that is comparable and what we’re working with for now.


The fluffy bench repository might not be enabled for public access btw. I can’t see it when I click on the link.


This is one of the major gripes that real estate sales agents have.

I’ve consistently heard complaints from customers that they don’t like it. At this point, map technology has come far enough that we could visualize thousands of points using Cesium on consumer-grade hardware so I don’t think it’s a performance issue either.

When it comes to overlapping dots, we generally solve this by displacing the dots minimally so that they’re all still in the same building or same area but visibly different locations.


This reminds me of DeepGrove AI’s ternary weight model: https://github.com/deepgrove-ai/Bonsai

Is there a particular difference that I might be missing here?


Location: Baltimore, MD, USA

Remote: Yes.

Willing to relocate: Yes, anywhere in the US.

Technologies: Languages[Python, Golang, TypeScript, Rust], Data Engineering Stack [DuckDB, Spark, Polars, FastAPI], GIS [Wails, React, Cesium], AI [HuggingFace, PydanticAI, PyTorch, vLLM, Transformers, OpenCV]

Resume: https://heltes.com/resume.pdf

Email: dat.adithya@gmail.com


The entire experiment just reminds me of Manna. We’re progressing a little too fast for comfort.

https://marshallbrain.com/manna1


Thank you for the reference, it's a fascinating read.

It would be good to highlight that this is fiction, though.


Hey Peter, I’m an F1 visa student and I’m trying to found a company with two Americans while being an Indian citizen. I plan on taking significant equity and will be part of the cofounders. Are there any significant hurdles that I should be on the lookout for here? I’d like to sign the founders agreement and found the company while I’m still pursuing my education but I’m not sure whether my visa status would throw any wrenches here.


It’s pretty much all you see nowadays on LinkedIn. Instagram is infected by AI videos that Sora generates while X has extremist views pushed up on a pedestal.


I suppose it’s only worth 256 words at a time right now. ;)

https://arxiv.org/abs/2010.11929


The CALM paper https://shaochenze.github.io/blog/2025/CALM/ says it is possible to compress 4 tokens in a single embedding, so... image = 4×256=1024 words > 1000 words. QED


2.4% relative error is not bad.


Reminds me of Babbage making allowance for meter.

"""

    ... it is said that he [Babbage] sent the following letter to Alfred, Lord Tennyson about a couplet in "The Vision of Sin":

         Every minute dies a man,
         Every minute one is born

    I need hardly point out to you that this calculation would tend to keep the sum total of the world's population in a state of perpetual equipoise, whereas it is a well-known fact that the said sum total is constantly on the increase. I would therefore take the liberty of suggesting that in the next edition of your excellent poem the erroneous calculation to which I refer should be corrected as follows:

         Every minute dies a man,
         And one and a sixteenth is born

    I may add that the exact figures are 1.167, but something must, of course, be conceded to the laws of metre.
"""

    Charles Babbage and his Calculating Engines


Shouldn't it be the other way around if the population is increasing? Every minute one is born = 1440 born/day, every minute and a sixteenth ~= 1335 dead/day for a net population increase of 105/day.


It means that in every minute, one and a sixteenth of a man is born.


Wouldn't "one and a sixth" be more accurate in both respects?


how do you decompress all those 4 words from one token?


Not from one token, from one embedding. Text contains a low amount of information: it is possible to compress a few token embeddings into a single tiken embedding.

The how is variable. The calm paper seems to have used a MLP to compress from and ND input (N embeddings of size D) into a single D embedding and other for decompress them back


The mechanism would be prediction (learnt during training), not decompression.

It's the same as LLMs being able to "decode" Base64, or work with sub-word tokens for that matter, it just learns to predict that:

<compressed representation> will be followed by (or preceded by) <decompressed representation>, or vice versa.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: