Second Brain

Index

Jan 12, 20261 min read

000 Zettelkasten - atomic notes on concepts and ideas
101 Literature - papers and paper notes

Recent Notes

Adaptation of Image Models to Video
Jan 11, 2026
Multi-View Foundation Models
Dec 17, 2025
- foundation_models
Into the Rabbit Hull - From Task-Relevant Concepts in DINO to Minkowski Geometry
Oct 08, 2025
Rethinking JEPA - Compute-Efficient Video SSL with Frozen Teachers
Sep 29, 2025
- video
- ssl
DINOv3
Aug 13, 2025
Back to the Features - DINO as a Foundation for Video World Models
Jul 25, 2025
EA-ViT - Efficient Adaptation for Elastic Vision Transformer
Jul 25, 2025
- efficient_vision
Franca - Nested Matryoshka Clustering for Scalable Visual Representation Learning
Jul 18, 2025
KV Cache Steering for Inducing Reasoning in Small Language Models
Jul 11, 2025
NeoBabel - A Multilingual Open Tower for Visual Generation
Jul 08, 2025
- multimodal
- llm
IntPhys 2 - Benchmarking Intuitive Physics Understanding In Complex Synthetic Environments
Jun 11, 2025
V-JEPA 2 - Self-Supervised Video Models Enable Understanding, Prediction and Planning
Jun 11, 2025
MoSiC - Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning
Jun 10, 2025
- computer_vision
- ssl
Advancing Video Self-Supervised Learning via Image Foundation Models
Jun 01, 2025
- computer_vision
Self-supervised learning of intertwined content and positional features for object detection
Jun 01, 2025
- vit
- ssl
Exploring Temporally-Aware Features for Point Tracking
Apr 20, 2025
- computer_vision
Learning from Streaming Video with Orthogonal Gradients
Apr 02, 2025
Principal Components Enable A New Language of Images
Mar 11, 2025
Deep Learning is Not So Mysterious or Different
Mar 03, 2025
Rate–Distortion–Perception Trade-Off in Information Theory, Generative Models, and Intelligent Communications
Feb 26, 2025
- information_theory
- dl_theory

Second Brain

Explorer

Index

Recent Notes

Adaptation of Image Models to Video

Multi-View Foundation Models

Into the Rabbit Hull - From Task-Relevant Concepts in DINO to Minkowski Geometry

Rethinking JEPA - Compute-Efficient Video SSL with Frozen Teachers

DINOv3

Back to the Features - DINO as a Foundation for Video World Models

EA-ViT - Efficient Adaptation for Elastic Vision Transformer

Franca - Nested Matryoshka Clustering for Scalable Visual Representation Learning

KV Cache Steering for Inducing Reasoning in Small Language Models

NeoBabel - A Multilingual Open Tower for Visual Generation

IntPhys 2 - Benchmarking Intuitive Physics Understanding In Complex Synthetic Environments

V-JEPA 2 - Self-Supervised Video Models Enable Understanding, Prediction and Planning

MoSiC - Optimal-Transport Motion Trajectory for Dense Self-Supervised Learning

Advancing Video Self-Supervised Learning via Image Foundation Models

Self-supervised learning of intertwined content and positional features for object detection

Exploring Temporally-Aware Features for Point Tracking

Learning from Streaming Video with Orthogonal Gradients

Principal Components Enable A New Language of Images

Deep Learning is Not So Mysterious or Different

Rate–Distortion–Perception Trade-Off in Information Theory, Generative Models, and Intelligent Communications

Graph View