eduKate SG

Category: Artificial Intelligence

Predictive Representation Learning | How Models Learn by Predicting Hidden Latent Structure Instead of Reconstructing Pixels
World-State Representation | How Models Turn Observations and Actions Into Predictive Latent State
Retrieval Representation | How Queries and Documents Become Sparse, Dense and Hybrid Search Signals
Expert Routing | How Mixture-of-Experts Models Decide Which Parameters Process Each Token
State-Space Representation | How Selective Recurrent State Carries Sequence History Without Full Attention
Latent Bottlenecks | How Models Compress Many Inputs Into a Small Working Representation
Uncertainty Representation | How Models Express Confidence, Ambiguity and When to Abstain
Context Sufficiency | When Retrieved Evidence Is Enough to Answer
Memory Representation | How Models Store, Compress, Forget and Retrieve Context
Set and Permutation Representation | How Models Work With Unordered Collections Without Inventing Sequence
Hierarchical Representation | How Local Tokens Become Sentences, Sections, Scenes and Global Memory
Token Interaction Topology | How Attention Masks, Local Windows and Global Tokens Shape Context
Positional Representation | How Models Know Where a Token Is
Adaptive Tokenisation | How Models Learn Which Pieces of Information Deserve Compute
Tokenisation-Free Models | What Happens When Models Read Bytes and Characters Directly?

Subscribe to
our newsletter