cd/entity/HHH· home entities HHH
grep -l @hhh /news/*.json | wc -l → 1

HHH

mentions 1 type Organization feed RSS

// recent coverage 1 mentions

16:21
2026-07-06
forum.effectivealtruism.org
ai-safety

Tie training can make DPO/RLHF-trained AIs generalize better

Researchers at ICML 2025 proved that AI models trained with DPO or RLHF inevitably learn spurious correlations, causing them to rely on non-causal features like verbosity or sycophancy instead of true…

// co-occurs with top 3 entities