Manacá-1B-Instruct-2: Reproducible Safety Alignment for a Native Brazilian-Portuguese LLM. Negative DPO Results, a Controlled Composition Scan, and a Silent Failure Mode of the LLM Judge

Manacá-1B-Instruct-2: Reproducible Safety Alignment for a Native Brazilian-Portuguese LLM. Negative DPO Results, a Controlled Composition Scan, and a Silent Failure Mode of the LLM Judge. It centres on AI Safety, and also names Benchmarks. Reported by arXiv. Bharat Hunt files it under AI Research and AI Models.

Written by Bharat Hunt from the headline and the coverage below. The original reporting is the source of truth.