ollin

6 ms·

1.
by ollin 2mo ago
This paper shows a nice integration of older winner-take-all ideas for learning K-modal generative models (see e.g. https://arxiv.org/abs/1612.00197 , https://arxiv.org/abs/2211.14286 ) into modern
2.
by ollin 2mo ago
The ExploitGym paper evaluated several frontier models on the bench and reported that "Different models find different exploits" [1], so it seems most plausible that the "test solutions directly from Hugging Face’s production
3.
by ollin 2mo ago
If the HuggingFace repo the agent broke into contains reference solution scripts for ExploitGym (i.e. for exploiting the vulnerabilities in the intended way), the agent can then run that reference code inside its original sandbox to retriev
4.
by ollin 2mo ago
This article seems fairly uninformative since, as others have pointed out, there's no visualization or comparison of the full emoji set and no link to see it. They just show a few example images and have some (AI-enhanced?) prose that
5.
by ollin 4mo ago
Yeah, that part should not be read literally; `x.cos().cos()` and `x1 = x.cos(); x2 = x1.cos()` both launch the same number of kernels (two in unfused/eager mode, one in fused/torch.compile, see this test notebook [1]). I think th
6.
by ollin 4mo ago
This post is a classic! Also recommended: Horace also gave a related talk (covering the high-level picture of modern ML Systems) at Jane Street in Dec 2024 https://www.youtube.com/watch?v=139UPjoq7Kw
7.
by ollin 4mo ago
Feet of Clay is one of my favorites in the series! It's surprising how literally the Discworld version of Golems corresponds to modern LLMs (and perhaps upcoming LLM-backed humanoids?). The Golems are brought to life by a slip of words
8.
by ollin 4mo ago
Specifically it looks like he's planning to extend the ideas from https://github.com/karpathy/autoresearch into a larger effort towards recursive training improvement [1]: > Excited to welcome Andrej to the Pre
9.
by ollin 4mo ago
This is cute and retro! But I think training only on GoldenEye undersells the concept a bit, since their world model inherits the N64-era graphics from GoldenEye, which automatically makes it look dated. If they retrained the same model on
10.
by ollin 4mo ago
Right now there is (AFAIK) no world model product booking any meaningful revenue. So there's a decent chance WMs turn out to have no long-term utility at all. However, there are a few promising markets, assuming WMs continue to get bet
11.
by ollin 4mo ago
This is a HW4 Tesla on FSD 14.3.2 trying to drive into a lake five days ago (a la The Office): https://www.reddit.com/r/TeslaFSD/comments/1t9rl2u/fsd_tried... , so I would not say Tesla has solved standin
12.
The Singularity Is the Friends We Made Along the Way(geohot.github.io)
2 points by ollin 4mo ago|0 comments
13.
by ollin 5mo ago
Most ONNX files are fp32, but the ONNX format actually allows fp16, int8, etc. as well (see onnx.proto for the full list of dtypes [1] - they even have fp8/fp4 these days!). I ended up switching over to fp16 ONNX models for my own web-
14.
by ollin 5mo ago
The source here is "CAISI Evaluation of DeepSeek V4 Pro" [1]; the US NIST ran their own benchmarks (including several internal ones) and reported the following table: | Domain | Benchmark | Model (
15.
by ollin 5mo ago
Hank Green has a video walking through how to use the timeline here https://www.youtube.com/watch?v=LyZE9VWJjDA . For me, the best experience was to click "Crew Photos Only" and then step through the photos chronol
16.
by ollin 5mo ago
For context, two days ago some users [1] discovered this sentence reiterated throughout the codex 5.5 system prompt [2]: > Never talk about goblins, gremlins, raccoons, trolls, ogres, pigeons, or other animals or creatures unless it is a
17.
by ollin 5mo ago
AFAIK Anthropic hasn't built any image or video generation tools yet, just text/code generation. OpenAI/Google/xAI all built image/video generation teams though so it may only be a matter of time.
18.
by ollin 5mo ago
- The OpenBSD one is 'TCP packets with invalid SACK options could crash the kernel' https://cdn.openbsd.org/pub/OpenBSD/patches/7.8/common/025_s... - One (patched) Linux kernel bug is &#x2
19.
by ollin 5mo ago
My impression was entirely the opposite; the unsolved subset of SWE-bench verified problems are memorizable (solutions are pulled from public GitHub repos) and the evaluators are often so brittle or disconnected from the problem statement
20.
by ollin 6mo ago
Here was the developer thread https://developer.apple.com/forums/thread/818403 I found with lots of other reports of "Unable to Verify App - An internet connection is required to verify the trust of the devel
21.
by ollin 7mo ago
The still photo (with 富士康科技 photoshopped out) is the second image of the "In Houston, workers assemble advanced AI servers" photo carousel https://www.apple.com/newsroom/images/2026/02/apple-acc
22.
by ollin 8mo ago
A lot of people mentioned this! The "dreamlike" comparison is common as well. In both cases, you have a network of neurons rendering an image approximating the real world :) so it sort of makes sense. Regarding the specific boilin
23.
by ollin 8mo ago
On a technical level, this looks like the same diffusion transformer world model design that was shown in the Genie 3 post (text/memory/d-pad input, video output, 60sec max context, 720p, sub-10FPS control latency due to 4-frame t
24.
by ollin 8mo ago
Yup, similar concepts! Just at two opposite extremes of the compute/scaling spectrum. - That forest trail world is ~5 million parameters, trained on 15 minutes of video, scoped to run on a five-year-old iPhone through a twenty-year old
25.
by ollin 8mo ago
Really great to see this released! Some interesting videos from early-access users: - https://youtu.be/15KtGNgpVnE?si=rgQ0PSRniRGcvN31&t=197 walking through various cities - https://x.com/fofrAI/sta
26.
by ollin 8mo ago
Z-Image is another open-weight image-generation model by Alibaba [1]. Z-Image Turbo was released around the same time as (non-Klein) FLUX.2 and received generally warmer community response [2] since Z-image Turbo was faster, also high-quali
27.
by ollin 8mo ago
https://madebyoll.in I write about on-device generative models (particularly world models). Past posts have been reasonably well-received on HN ( https://news.ycombinator.com/from?site=madebyoll.in ).
28.
by ollin 11mo ago
Yeah the issue reads as if someone asked Claude Code "find the most serious performance issue in the VSCode rendering loop" and then copied the response directly into GitHub (without profiling or testing anything).
29.
by ollin 1y ago
Notes on my experience: - Infra/systems : I was able to connect to a server within a minute or two. Once connected, the displayed RTT (roundtrip time?) was around 70ms but actual control-to-action latency was still around ~600-700ms v
30.
by ollin 1y ago
Regarding latency, I found a live video of gameplay here [1] and it looks like closer to 1.1s keypress-to-photon latency (33 frames @ 30fps) based on when the onscreen keys start lighting up vs when the camera starts moving. This writeup [2

More ›