AI Wiki — Knowledge Base

from Write & Learn (thx @MilksandMatcha @philipkiely @swyx), my most important takeaway, 𝐛𝐮𝐢𝐥𝐝 𝐚 𝐝𝐞𝐧𝐬𝐞 𝐫𝐞𝐰𝐚𝐫𝐝 𝐟𝐨𝐫 𝐲𝐨𝐮…

from Write & Learn (thx @MilksandMatcha @philipkiely @swyx), my most important takeaway, 𝐛𝐮𝐢𝐥𝐝 𝐚 𝐝𝐞𝐧𝐬𝐞 𝐫𝐞𝐰𝐚𝐫𝐝 𝐟𝐨𝐫 𝐲𝐨𝐮𝐫 𝐰𝐫𝐢𝐭𝐢𝐧𝐠. fun fact. most people know @swyx as the famous podcaster, the AIE godfat

tutorialsswyx--x

26 Jun 2026

Search: “tutorials”

from Write & Learn (thx @MilksandMatcha @philipkiely @swyx), my most important takeaway, 𝐛𝐮𝐢𝐥𝐝 𝐚 𝐝𝐞𝐧𝐬𝐞 𝐫𝐞𝐰𝐚𝐫𝐝 𝐟𝐨𝐫 𝐲𝐨𝐮…

Generating a Paracosm for Training-Free Zero-Shot Composed Image Retrieval

How to evaluate clustering with ground truth?

Modeling Local, Global, and Cross-Modal Context in Multimodal 3D MRI

Neural Architecture Search for Generative Adversarial Networks: A Comprehensive Review and Critical Analysis

Neural Texture Compression using Hypernetworks

Neural Voxel Dynamics: Learning Implicit 3D Physics via Volumetric Feature Advection

OpenAI says 97.9% of its employees are now using Codex, up from ~40% in August 2025; non-developer usage of Codex has risen 137x for individual users (Thomas Claburn/The Register)

Physics-guided Convolutional Neural Network for Domain Growth Prediction in Systems with Conserved Kinetics

Play2Perfect: What Matters in Dexterous Play Pretraining for Precise Assembly?

Power Couple? AI Growth and Renewable Energy Investment

SatSplatDiff: Geometry-preserving generative refinement for high-fidelity satellite Gaussian Splatting

Scalable Behavior Cloning with Open Data, Training, and Evaluation

Simple Blender animation → Seedance 2.0 → stylized AI video. This workflow uses basic 3D motion as a guide, then applies the look and detail…

Structure Before Collapse: Transient semantic geometry in next-token prediction

Theory of the Frequency Principle for General Deep Neural Networks

this is too late for AIE worlds fair but heres the how to book talks in the first place https://x.com/swyx/status/2069985832166154339?s=46

Transformers are better at copying, while RNNs are better at modeling 'meaning-bearing words—the nouns, verbs, & adjectives that say what a …

Adapt Only When It Pays: Budgeted Decision-Loss Priority for Delayed Online Time-Series Adaptation

An Integrated Hardware-Software Design for Low-Data Spatial Defect Detection in Robotic Visual Inspection with Hybrid Optoelectronic Neural Networks

Breaking Data Symmetry is Needed For Generalization in Feature Learning Kernels

Compositional Behavioral Semantics for State Abstraction in Reinforcement Learning

Cross-Modality Structural Guidance in 3D Latent Diffusion for Robust FLAIR Super-Resolution

Did Models Learn Sufficiently? Attribution-Guided Training via Subset-Selected Counterfactual Augmentation

Does Translation-Enhanced Speech Encoder Pre-training Affect Speech LLMs?

ESTANet: Efficient Online Error Detection in Procedural Videos via Prediction Inconsistency

Facebook’s Creator Studio has been revived as an AI companion app

H-Adapter: Pose-Robust Hairstyle Transfer via Attention-Derived, Source-Aligned Hair Masks

How Complexity Contributes to Learning Opacity in Machine Learning

I think part of what makes this account’s work so satisfying is they are doing a form of pattern recognition that is computerlike in many wa…

Implementing super resolution by deploying SeedVR2 on Amazon SageMaker AI

🚨 In a surprise to NOBODY, a study shows that generative AI use raises homework scores, but substantially reduces learning. My takeaways fo…

Interpretable Concept-Guided Polynomial Tabular Kolmogorov-Arnold Network for EEG-Based Mild Cognitive Impairment Detection

Knowledge Cascade: Reverse Knowledge Distillation on Nonparametric Multivariate Functional Estimation

Learning task-specific subspaces via interventional post-training of speech foundation models

Lifelong In-Context Learning with Transformers Requires Parametric Forms of Attention

LLM Program Optimization via Retrieval Augmented Search

lots of folks prepping talks next week (congrats!). Some thoughts from RLing on thousands of hours of engineer- and researcher- focused talk…

MJEPA: A Simple and Scalable Joint-Embedding Predictive Architecture for Audio-Visual Learning

Multifidelity-Augmented Gaussian Process Inputs for Surrogate Modeling from Scarce Data

obvious in retrospect but I had no idea there was a black market for tokens

Or people could learn their lesson? Naah 🙈

Stabilizing black-box algorithms through task-oriented randomization

Taxonomy of Risks on Automated Fact-Checking Systems Considering its Propagation

We're sharing new research on how models hack public benchmarks. The latest models, including Opus 4.8 and Composer 2.5, learn to retrieve s…

What happens when you shift more of the context layer into the model weights themselves? Engram is building a neolab focused on memory and c…

What's in an Earth Embedding? An Explainability Analysis of Location Encoders

While everyone’s displacing jobs, we’re creating them. A Hero for every family. @hellotohera’s favorite friend 83 year old Stanley Zareff dr…