[ Novel-Research ]

Discover insights, tutorials, and thoughts on technology, homelab, and development.

Dec 28, 2025 5 min read

Beyond Benchmark Gaming: Multi-Model Consensus for Genuinely Capable AI

Using agreement among frontier models (Claude, GPT-5, Gemini) as a training signal to build AI that's genuinely capable, not just benchmark-optimized

ai-training consensus goodhart

Dec 16, 2025 5 min read

A novel discovery: hidden state inversion quality predicts model capability, enabling self-improving systems without external feedback

ai-training self-improvement transformers