Stories by zerojames

Step Differences in Instructional Video

Ask HN: How do you sift through new research?

MoDE: CLIP Data Experts via Clustering

TorchGeo: Datasets and pre-trained models for geospatial data

FxEmojis: A friendly emoji set from Mozilla

"Testing on the Toilet" (2007)

Diarizers: Fine-tune speaker diarization models

Removing Reflections from Raw Photos

Maxtext: A simple, performant and scalable Jax LLM

MobileCLIP: Fast Image-Text Models Through Multi-Modal Reinforced Training

Overcoming the Pitfalls of Vision-Language Model Finetuning – OOD Generalization

American flag sort

VGGSfM: Visual Geometry Grounded Deep Structure from Motion

VSCode Essentials eBook

OpenAI Japan

The Joy of Computing

The Joy of Computing

Show HN: Train Station Pianos

Using random boolean networks in compression

Accessible Numbers

5
6
7
8
9
10
11
12
13
14