Megadose AI progress, ranked and analyzed.

Geometry of Values: Task Vector Composition for Ethical Preference Alignment in Language Models

· HF Daily Papers ·
The paper tests whether ethical preferences in language models can be isolated as editable directions, rather than treated as broad fine-tuning behavior.

The authors introduce a 12,000-instance dilemma dataset spanning conflicts among Honesty, Justice, and Autonomy, translated into Hindi, Arabic, Spanish, and Chinese. In their benchmark, GPT-5-mini favors Honesty over Autonomy across all five languages when no policy is supplied. Llama-3.2-1/3B shows strong first-option bias, which plain fine-tuning and DPO both reduce while lifting accuracy above 98%. The paper then uses orthogonalized task vectors to separate value preference from general instruction-following and flip a model toward the opposite stance. HF Daily Papers' note

score 4

Categories: Research