freedmand

7 ms·

1.
Anatomy of an Autonomous Attack: 5 Alarming A.I. Capabilities(nytimes.com)
2 points by freedmand 25d ago|0 comments
2.
by freedmand 1y ago
Gift link (no subscription needed): https://www.nytimes.com/2025/08/19/business/chatgpt-gpt-5-ba...
3.
The Day ChatGPT Went Cold(nytimes.com)
4 points by freedmand 1y ago|2 comments
4.
The Chatbot Updated. Users Lost a Friend(nytimes.com)
2 points by freedmand 1y ago|0 comments
5.
Chatbots can go into a delusional spiral. Here’s how it happens.(nytimes.com)
2 points by freedmand 1y ago|1 comments
6.
by freedmand 1y ago
Here's a gift link: https://www.nytimes.com/interactive/2025/03/26/business/ai-s...
7.
by freedmand 2y ago
Do you think that there are security or data exfiltration risks with their approach, e.g. putting classified documents/emails into private company's LLMs to summarize information or identify certain ideological motivations? I don&
8.
by freedmand 2y ago
This will probably be my last run (2:49): https://youtu.be/zIyZJjyPy8U Kudos on a great game! If you ever expand or open source this, let me know. The mechanics and design are great and could really blossom into something f
9.
by freedmand 2y ago
Oh wow, looks like someone uploaded a 2:55 yesterday: https://www.youtube.com/watch?v=G6yExxPR1vI
10.
by freedmand 2y ago
I'm by no means a speedrunner but decided to record my runs in case I got a sub-3. Just uploaded the 2:57: https://youtu.be/0THaZ61sCmg I challenge someone to beat it!
11.
by freedmand 2y ago
I got 2:57 — required a near-perfect run but still room for improvement
12.
by freedmand 2y ago
I got 2:57 after discovering a few shortcuts
13.
by freedmand 2y ago
Can't say it'll solve all your problems, but try Florence-2. It's worked well on some handwritten documents for me when all the text is a relatively uniform size. Model: https://huggingface.co/microsoft/F
14.
by freedmand 2y ago
If you play ten note chords — one for each finger — in quick succession, that can rack up a lot of inserts in short time period (say, medium-worst case, 100Hz, for playing a chord like that five times per second, counting both “on” and “off
15.
by freedmand 2y ago
Would you expect SQLite performance to degrade per insert once a table is very large (in a way that, say, an append-only log file wouldn’t)?
16.
by freedmand 2y ago
I recently had the idea to record every note coming out of my digital piano in real-time. That way if I come up with a good idea when noodling around I don’t have to hope I can remember it later. I was debating what storage layer to use and
17.
by freedmand 2y ago
Agreed. Tesseract is not able to handle handwriting or text that is distorted well, e.g. colored text over an image background — to the point that it would hurt any downstream LLM trying to make sense of the contents. It won’t even pick out
18.
by freedmand 2y ago
I just found this on HuggingFace: https://huggingface.co/mistralai/Mixtral-8x22B-Instruct-v0.1
19.
by freedmand 2y ago
The article from Bloomberg never said "racist" — it said tests revealed racial bias. The "racist" term is from the title refutation piece.
20.
by freedmand 3y ago
Each box is a torch module[1], which can technically encapsulate any amount of computation or contain sub-modules. The smallest boxes or leaf nodes are those that contain no sub-modules. They can still run an arbitrary number of functions d
21.
Show HN: Library to visualize and inspect PyTorch models(github.com)
4 points by freedmand 3y ago|2 comments
22.
by freedmand 3y ago
I wonder how people’s learning habits will change with AI tools like GitHub Copilot. I used it for several months but randomly got logged out and am now finding myself valuing the slight inconvenience of looking up official documentation (a
23.
by freedmand 3y ago
Semantra! Shared it yesterday on HN https://github.com/freedmand/semantra
24.
by freedmand 3y ago
Hi HN, I’m excited to present Semantra, a tool and search UI where you can easily embed/query your own documents semantically (by meaning). Semantic embedding models are lightweight and can run on consumer grade CPUs. This project prov
25.
Show HN: Semantra – local, private semantic search engine for your files(github.com)
4 points by freedmand 3y ago|1 comments
26.
by freedmand 3y ago
I don’t fully understand the fascination with retrieval augmented generation. The retrieval part is already really good and computationally inexpensive — why not just pass the semantic search results to the user in a pleasant interface and
27.
by freedmand 3y ago
I don't disagree with anything you said. I'm saying more that if the compression algorithm is benchmarking against "Alice in Wonderland" and has consumed the entirety of "Alice in Wonderland" in training the LL
28.
by freedmand 3y ago
Was it run on any text that was not feasibly training data for the LLMs? It wouldn't be a fair comparison otherwise
29.
by freedmand 3y ago
I have proof-of-concepted[1] building an entirely static website offering client-side semantic search using transformers.js[2] and Semantra*[3] with MiniLM-L12[4]. It's usable on fast connections and decent CPUs but is not delightfully
30.
by freedmand 3y ago
Never shared this before, but I had a phase where I built satirical landing pages and of them was “Words: The programming language for ideas” https://readymag.com/dylan/words/

More ›