# VISPA: Pluralistic Alignment via Automatic Value Selection and Activation

arXiv: https://arxiv.org/abs/2601.12758 · HTML: https://mlsanity.com/paper/2601.12758/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 28 of its 52 references that are in the map.
- First posted on arXiv 19 Jan 2026.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Human values and AI feedback alignment](https://mlsanity.com/alignment/human-values-and-ai-feedback-alignment/)

## Builds on (most cited first, 28 of 28)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://mlsanity.com/paper/2006.03654/) (2020)
- [Steering Llama 2 via Contrastive Activation Addition](https://mlsanity.com/paper/2312.06681/) (2023)
- [Large Language Models](https://mlsanity.com/paper/2307.05782/) (2023)
- [A Comprehensive Capability Analysis of GPT-3 and GPT-3.5 Series Models](https://mlsanity.com/paper/2303.10420/) (2023)
- [Get Your Vitamin C! Robust Fact Verification with Contrastive Evidence](https://mlsanity.com/paper/2103.08541/) (2021)
- [Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalization](https://mlsanity.com/paper/2406.01171/) (2024)
- [Sociotechnical Safety Evaluation of Generative AI Systems](https://mlsanity.com/paper/2310.11986/) (2023)
- [MaxMin-RLHF: Alignment with Diverse Human Preferences](https://mlsanity.com/paper/2402.08925/) (2024)
- [Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration](https://mlsanity.com/paper/2406.15951/) (2024)
- [On the Algorithmic Bias of Aligning Large Language Models with RLHF: Preference Collapse and Matching Regularization](https://mlsanity.com/paper/2405.16455/) (2024)
- [Word Embeddings Are Steers for Language Models](https://mlsanity.com/paper/2305.12798/) (2023)
- [Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaboration](https://mlsanity.com/paper/2404.12715/) (2024)
- [Massively Multi-Cultural Knowledge Acquisition & LM Benchmarking](https://mlsanity.com/paper/2402.09369/) (2024)
- [Controlling Large Language Models Through Concept Activation Vectors](https://mlsanity.com/paper/2501.05764/) (2025)
- [An Evaluation of Cultural Value Alignment in LLM](https://mlsanity.com/paper/2504.08863/) (2025)
- [Diverse Preference Learning for Capabilities and Alignment](https://mlsanity.com/paper/2511.08594/) (2025)
- [Exploring Multilingual Concepts of Human Value in Large Language Models: Is Value Alignment Consistent, Transferable and Controllable across Languages?](https://mlsanity.com/paper/2402.18120/) (2024)
- [VITAL: A New Dataset for Benchmarking Pluralistic Alignment in Healthcare](https://mlsanity.com/paper/2502.13775/) (2025)
- [Internal Value Alignment in Large Language Models through Controlled Value Vector Activation](https://mlsanity.com/paper/2507.11316/) (2025)
- [tasksource: A Dataset Harmonization Framework for Streamlined NLP Multi-Task Learning and Evaluation](https://mlsanity.com/paper/2301.05948/) (2023)
- [Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective](https://mlsanity.com/paper/2501.00581/) (2024)
- [Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behavior](https://mlsanity.com/paper/2511.14476/) (2025)
- [Steering LLMs for Formal Theorem Proving](https://mlsanity.com/paper/2502.15507/) (2025)
- [Pluralistic Alignment for Healthcare: A Role-Driven Framework](https://mlsanity.com/paper/2509.10685/) (2025)

## Built on by (most cited first, 1 of 1)

- [Adaptive Multi-Value Control in LLMs via Causal Activation Steering](https://mlsanity.com/paper/2609.30405/) (2026)

## Abstract (arXiv)

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.