Viewing profile — ollin

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/2211.14…

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 solu…

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 ru…

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 …

[dead]

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/torc…

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/…

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 humanoi…

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]: > Excite…

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 automatical…

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 …

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 …

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 ende…

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 | Be…

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 th…

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, ogr…

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…

- 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…

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 o…

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 v…

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/202…

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 …

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…

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 …