VISPA: Pluralistic Alignment via Automatic Value Selection and Activation

arXiv:2601.12758v2 Announce Type: replace-cross Abstract: As large language models are increasingly used in high-stakes domains, it is essential that their outputs reflect not average} human preference, rather range of varying perspectives. Achieving such pluralism, however, remains challenging. Existing approaches consider limited values or rely on prompt-level interventions, lacking value control and representation. To address this, we introduce VISPA, a training-free pluralistic alignment framework, that enables direct control over value expression by dynamic selection and internal model activation steering. Across extensive empirical studies spanning multiple models and evaluation settings, we show VISPA is performant across all pluralistic alignment modes in healthcare and beyond. Further analysis reveals VISPA is adaptable with different steering initiations, model, and/or values. These results suggest that pluralistic alignment can be achieved through internal activation mechanisms, offering a scalable path toward language models that serves all.

Related

• PICACO: Pluralistic In-Context Value Alignment of LLMs via Total Correlation Optimization
• Culturally Grounded Personas in Large Language Models: Characterization and Alignment with Socio-Psychological Value Frameworks
• VALUEFLOW: Toward Pluralistic and Steerable Value-based Alignment in Large Language Models
• Demographic Pluralism: Inference-Time Modeling of Pluralistic Human Preference Distributions

Source: arXiv cs.AI | 2026-10-02