Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
gok
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
31.
▲
Nvidia-backed Starcloud trains first AI model in space, orbital data centers
(cnbc.com)
8 points
by
gok
10mo ago
|
0 comments
32.
▲
This Century, Child Mortality Is Likely to Rise
(time.com)
12 points
by
gok
10mo ago
|
0 comments
33.
▲
Xzone Malloc
(github.com)
2 points
by
gok
10mo ago
|
0 comments
34.
▲
Meta prevails in historic FTC antitrust case
(apnews.com)
7 points
by
gok
11mo ago
|
0 comments
35.
▲
Empire of AI is wildly misleading on AI water use
(andymasley.substack.com)
2 points
by
gok
11mo ago
|
1 comments
36.
▲
Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
(github.com)
5 points
by
gok
11mo ago
|
0 comments
37.
▲
Arc Prize Verified
(arcprize.org)
3 points
by
gok
11mo ago
|
0 comments
38.
▲
My jury stood up to Trump's DOJ and acquitted me of assault on an FBI agent
(msnbc.com)
35 points
by
gok
11mo ago
|
1 comments
39.
▲
The Swift SDK for Android
(swift.org)
696 points
by
gok
1y ago
|
280 comments
40.
▲
by
gok
1y ago
> While they provide data privacy, our research shows their weaker reasoning and alignment capabilities make them easier targets for sabotage. If you are using any LLM's reasoning ability as a security boundary, something is deeply,
41.
▲
FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion LMs
(machinelearning.apple.com)
2 points
by
gok
1y ago
|
0 comments
42.
▲
Secure your app with Memory Integrity Enforcement [video]
(youtube.com)
1 points
by
gok
1y ago
|
0 comments
43.
▲
by
gok
1y ago
The real lesson here should be that doing crazy shit like swizzling the program counter in a signal handler and writing your own assembler is not a good idea.
44.
▲
Continuously Augmented Discrete Diffusion Model
(arxiv.org)
4 points
by
gok
1y ago
|
0 comments
45.
▲
AppMigrationKit
(developer.apple.com)
2 points
by
gok
1y ago
|
0 comments
46.
▲
Anduril and Palantir battlefield communication system has flaws, Army memo says
(cnbc.com)
221 points
by
gok
1y ago
|
92 comments
47.
▲
RISC-V Conditional Moves
(corsix.org)
79 points
by
gok
1y ago
|
71 comments
48.
▲
Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models
(arxiv.org)
4 points
by
gok
1y ago
|
1 comments
49.
▲
SimpleFold: Folding Proteins Is Simpler Than You Think
(arxiv.org)
2 points
by
gok
1y ago
|
0 comments
50.
▲
America's Next Top Model Context Protocol Server [video]
(youtube.com)
7 points
by
gok
1y ago
|
0 comments
51.
▲
Stephen Miller's Quota Likely Drove Korean Arrests in Immigration Raid
(forbes.com)
91 points
by
gok
1y ago
|
80 comments
52.
▲
by
gok
1y ago
tldr it's somewhat comparable to an A100, which was released in May 2020.
53.
▲
Implement Primary Support for MTE
(github.com)
2 points
by
gok
1y ago
|
0 comments
54.
▲
Instruction-Following Pruning for Large Language Models
(arxiv.org)
5 points
by
gok
1y ago
|
0 comments
55.
▲
by
gok
1y ago
This has been done in fact: https://doi.org/10.1016/j.sysarc.2024.103102
56.
▲
by
gok
1y ago
How hard would it be to add generational collection?
57.
▲
FastVLM WebGPU Demo
(huggingface.co)
3 points
by
gok
1y ago
|
0 comments
58.
▲
Path Guiding Surfaces and Volumes in Disney's Hyperion Renderer: A Case Study
(yiningkarlli.com)
2 points
by
gok
1y ago
|
0 comments
59.
▲
by
gok
1y ago
This is like asking a human how many pixels appears in the word "blueberry".
60.
▲
by
gok
1y ago
So this 4B dense model gets very similar performance to the 30B MoE variant with 7.5x smaller footprint.
More ›