jbarrow

6 ms·

1.
DeltaNet Explained (Part 1)(sustcsonglin.github.io)
2 points by jbarrow 21d ago|0 comments
2.
Show HN: Locus-v1, a dataset of 2.2M local laws and ordinances(huggingface.co)
2 points by jbarrow 3mo ago|0 comments
3.
by jbarrow 3mo ago
I'm always glad to see more multi-page work in VLM-based OCR. Especially single-pass. One of the few other multi-page papers from recently, MinerU-Popo, treats fixing up multi-page outputs as a post-processing correction step ( https:&
4.
by jbarrow 4mo ago
Very impressed with how much the Gemma ecosystem has advanced just this week. Gemma 12B, multitoken prediction, and official quants released. Feels like Google is putting real effort into this string of releases, and I'm very excited t
5.
Building a 1-Outlet, 4-GPU Workstation(jbarrow.ai)
2 points by jbarrow 4mo ago|0 comments
6.
Agents Have (Information) Needs(jbarrow.ai)
2 points by jbarrow 4mo ago|0 comments
7.
by jbarrow 6mo ago
Very cool to see a company pushing what's possible with (relatively) tiny models! A 350M parameter trained on 28T tokens that, from the benchmarks, is competitive with Qwen3.5-0.8B. Comparing the architecture to Qwen3.5, it seems: - fe
8.
LFM2.5-350M: No Size Left Behind(liquid.ai)
3 points by jbarrow 6mo ago|1 comments
9.
by jbarrow 6mo ago
Shared this because I was having fun thinking through floating point numbers the other day. I worked through what fp6 (e3m2) would look like, doing manual additions and multiplications, showing cases where the operations are non-associative
10.
What every computer scientist should know about floating-point arithmetic (1991) [pdf](itu.dk)
125 points by jbarrow 6mo ago|56 comments
11.
by jbarrow 6mo ago
I've been noticing a _lot_ more AI-generated/edited content of late, both comments and stories. It's gotten to the point that I spend a lot less time on HN than I used to, and if it continues to get worse I expect I'll q
12.
by jbarrow 7mo ago
Watsi is incredibly inspiring! I’ve been a monthly donor since ~the beginning when I was just an undergraduate, and I still read the stories and emails I receive. I’m glad that you opted for the steady growth path, and that you’ve made it a
13.
by jbarrow 8mo ago
The whole thing feels AI written, generated from the codebase.* *this is incorrect per the author’s response, my apologies. For instance, it goes into (nano)vLLM internals and doesn’t mention PagedAttention once (one of the core ideas that
14.
by jbarrow 11mo ago
Super interesting. Would you be willing to try the Python package ( https://github.com/jbarrow/commonforms ) or share the PDFs? For the non-ONNX models there are some inference tricks that generally improve performance,
15.
by jbarrow 11mo ago
Hey, Benjamin, thanks for the attribution! Happy to field any questions HN users have. It's really gratifying to see people building on the work, and I love that it's possible to do browser-side/on-device.
16.
by jbarrow 11mo ago
Woah, did not realize that, haha. Let me know if it works well!
17.
by jbarrow 11mo ago
Training ML models for PDF forms. You can try out what I’ve got so far with this service that automatically detects where fields should go and makes PDFs fillable: https://detect.semanticdocs.org/ Code and models are at: h
18.
by jbarrow 1y ago
Existing “auto-fillable” tools are pretty lackluster in my experience. CommonForms is tooling that can automatically detect form fields in PDFs and turn those PDFs into fillable documents. The dataset is ~500k form pages pulled from Common
19.
Show HN: CommonForms – open models to auto-detect PDF form fields(github.com)
1 points by jbarrow 1y ago|1 comments
20.
by jbarrow 1y ago
I’m personally a huge fan of Modal, and have been using their serverless scale-to-zero GPUs for a while. We’ve seen some nice cost reductions from using them, while also being able to scale WAY UP when needed. All with minimal development e
21.
by jbarrow 1y ago
If you enjoyed this essay, you should check out the author’s current project, Dynamicland[1]. It is a wonderful expression of what computing and interaction could be. Even the project website — navigating a physical shelf, and every part is
22.
by jbarrow 1y ago
Editing text in PDFs is _really_ hard compared to other document formats because most PDFs don't really encode the "physics" of the document. I.e. there isn't a notion of a "text block with word wrapping," it&#
23.
by jbarrow 1y ago
Wonderful! Inserted form-fields show up in Preview and Acrobat, which is not a trivial task. I run a little AI-powered tool that automatically figures out where form fields should go ( https://detect.penpusher.app ) and robustly a
24.
by jbarrow 2y ago
I'm sorry I missed this earlier, but I absolutely believe that it could do that. Do you have any pointers to PDF forms that work well or don't work well with screen readers? I'd be happy to take a look, and see if I can impro
25.
Show HN: Automatically Turn PDFs into Fillable Forms(detect.penpusher.app)
3 points by jbarrow 2y ago|2 comments
26.
by jbarrow 2y ago
> Unfortunately Gemini really seems to struggle on this, and no matter how we tried prompting it, it would generate wildly inaccurate bounding boxes Qwen2.5 VL was trained on a special HTML format for doing OCR with bounding boxes. [1] T
27.
by jbarrow 2y ago
Mostly because OpenAI's vision offerings aren't particularly compelling: - 4o can't really do localization, and ime is worse than Gemini 2.0 and Qwen2.5 at document tasks - 4o mini isn't cheaper than 4o for images becaus
28.
by jbarrow 2y ago
I've been very impressed by Gemini 2.0 Flash for multimodal tasks, including object detection and localization[1], plus document tasks. But the 15 requests per minute limit was a severe limiter while it was experimental. I'm reall
29.
by jbarrow 2y ago
Loved the one in Kansas City! There are some great, thematically-similar museums in other countries as well, if you ever find yourself there: - the Vasa in Stockholm, Sweden is a ship dredged from the harbor and stabilized, sank in 1628 - t
30.
by jbarrow 2y ago
> There will be more cloud-based products turned to bricks by manufacturers that go bankrupt or simply stop caring. This one feels like a gimme. The recent Garmin outage that partially bricked the Connect app was a bit of a surprise; so

More ›