Notes

This is where I keep my notes.

Posts

• Explain, Predict and Control (Teuber 1974)
• The Phase Spectrum
• MaCo - Feature Visualization for Deeper Networks
• Optically Pumped Magnetometers
• Video Generation Models
• Direct Preference Optimization for Diffusion Models
• Singular Value Decomposition
• Matrices
• DreamSim
• MindEye: fMRI-to-Image with Contrastive Learning
• BIMT: Brain Inspired Modular Training
• AudioCLIP: Extending CLIP to Image, Text and Audio
• THINGS: a multimodal dataset for investigating object representations in thehuman brain
• Open Vocabulary EEG-To-Text Decoding
• Brain2Image: Converting Brain Signals into Images
• Neural Taskonomy - using encodings from vision models to understand the human brain
• Locating and Editing Factual Knowledge in GPTs
• Language Models (mostly) Know What they know
• Order From Chaos (Part 1): Diffusion for image synthesis explained in simple words
• Order From Chaos (Part 2): Diffusion for image synthesis explained in code and a little bit of math
• CLIP-Fields: Weakly Supervised Semantic Fields for Robotic Memory
• CLIPPO: Image and Language understanding from pixels only
• VPT: Video Pre-Training on Minecraft
• Brain to Text Communication via handwriting
• CraftAssist - LLMs on minecraft (?)
• Watching artificial intelligence through the lens of CogSci
• Towards a New Science of Common Sense
• Parti - Scaling LLMs for Text to Image tasks
• The Third Wave (?)
• Transformer intrepretability beyond attention
• Self-Consistency Improves Chain of Thought Reasoning in LMs
• LOST - Localizing objects with self supervised transformers
• Knowledge Neurons
• Interpreting LMs with contrastive explanations
• Generic Attention-model Explainability
• Attention Rollout