
Machine Learning
ByteDance’s Lance: A 3‑Billion‑Parameter Unified Multimodal Model
5/22/2026

Machine Learning
CODA: Optimizing Transformer Performance by Redefining Kernel Abstractions
5/22/2026

LLMs
Multi-Stream LLMs: Parallelising Thought, Input and Output to Unblock Language Agents
5/21/2026

LLMs
δ‑mem: A Tiny Online Memory That Boosts Large Language Models Without Expanding Context
5/16/2026

Machine Learning
Building Your Own LLM From Scratch: A Hands-On Workshop for Understanding AI Fundamentals
5/5/2026
Machine Learning
LoGeR: Breaking the Quadratic Barrier in Long-Video 3D Reconstruction
3/10/2026
Machine Learning
MicroGPT Emerges as Blueprint for Ultra-Compact Transformer Models
2/16/2026
AI
The Persistent Gap: Why Transformer-Based AI Falls Short of Human Cognition
2/16/2026

LLMs
microgpt: A 200-Line Pure Python GPT Implementation
2/13/2026

AI
Constant-Cost Self-Attention Breakthrough Could Revolutionize AI Scaling
2/4/2026

Machine Learning
Foundation Models for Ranking: Challenges, Successes, and Lessons Learned
1/28/2026

Machine Learning
Tauformer: Topological Transformer Shows Promise with Efficient Attention Mechanism
1/18/2026

AI
Sparrow-1: Building Conversational Timing That Feels Human
1/15/2026