📰 Dev.to · felipe muniz

Geometric Alignment: Can Curved Embedding Spaces Make AI Safer?

DRM-Transformer — Intrinsic Geometry for Structural Alignment

We Found Toroidal Topology Emerging in a Neural Epistemic Manifold

Encoding Human Values as Geometry: The Gravitational Objective

How to Measure Whether Your Model's Uncertainty Space Is Flat or Curved

The Three Doors Problem: Why RLHF Systems Slide Toward Autonomy

The Manifold Game

I trained a 354M LLM alone and it outperforms GPT-2 Medium in epistemic calibration

The Semantic Gatekeeping Problem in AI Research

ATIC Doesn't Train. It Thinks. — How a Brazilian Developer Hit #1 on LiveBench Without Touching a Single Weight

How I Built a System to Detect When LLMs Don't Know Something