mlsanityData to 14 Oct 2026

VISPA: Pluralistic Alignment via Automatic Value Selection and Activation

arXiv 2601.12758 ↗First posted 19 Jan 2026cs.CL cs.AI cs.LGcited by 1

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

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.

Key facts. Cited by 1 paper 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.

Builds on

28 of 52 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Operationalizing Pluralistic Values in Large Language Model Alignment Reveals Trade-offs in Safety, Inclusivity, and Model Behavior18 Nov
  2. 2025Internal Value Alignment in Large Language Models through Controlled Value Vector Activation15 Jul
  3. 2024Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective31 Dec
  4. 2024Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration22 Jun
  5. 2024Exploring Multilingual Concepts of Human Value in Large Language Models: Is Value Alignment Consistent, Transferable and Controllable across Languages?28 Feb

Other references · 23

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  4. 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,898
  5. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,181
  6. 2023Steering Llama 2 via Contrastive Activation Additioncited by 1,085
  7. 2023Large Language Modelscited by 780
  8. 2023A Comprehensive Capability Analysis of GPT-3 and GPT-3.5 Series Modelscited by 246
  9. 2021Get Your Vitamin C! Robust Fact Verification with Contrastive Evidencecited by 225
  10. 2024Two Tales of Persona in LLMs: A Survey of Role-Playing and Personalizationcited by 208
  11. 2023Sociotechnical Safety Evaluation of Generative AI Systemscited by 157
  12. 2024MaxMin-RLHF: Alignment with Diverse Human Preferencescited by 123
11 more
  1. 2024On the Algorithmic Bias of Aligning Large Language Models with RLHF: Preference Collapse and Matching Regularizationcited by 72
  2. 2023Word Embeddings Are Steers for Language Modelscited by 70
  3. 2024Ensemble Learning for Heterogeneous Large Language Models with Deep Parallel Collaborationcited by 67
  4. 2024Massively Multi-Cultural Knowledge Acquisition & LM Benchmarkingcited by 63
  5. 2025Controlling Large Language Models Through Concept Activation Vectorscited by 28
  6. 2025An Evaluation of Cultural Value Alignment in LLMcited by 28
  7. 2025Diverse Preference Learning for Capabilities and Alignmentcited by 25
  8. 2025VITAL: A New Dataset for Benchmarking Pluralistic Alignment in Healthcarecited by 17
  9. 2023tasksource: A Dataset Harmonization Framework for Streamlined NLP Multi-Task Learning and Evaluationcited by 11
  10. 2025Steering LLMs for Formal Theorem Provingcited by 4
  11. 2025Pluralistic Alignment for Healthcare: A Role-Driven Frameworkcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Adaptive Multi-Value Control in LLMs via Causal Activation Steeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024A Roadmap to Pluralistic Alignment
  2. 2026VALUEFLOW: Toward Pluralistic and Steerable Value-based Alignment in Large Language Models
  3. 2025Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression
  4. 2026PLURAL: A Global Dataset for Value Alignment
  5. 2025PICACO: Pluralistic In-Context Value Alignment of LLMs via Total Correlation Optimization
  6. 2025DiverValue-Bench: A Benchmark and Fine-Tuning Framework for Aligning Large Language Models with Diverse Human Values
  7. 2023Heterogeneous Value Alignment Evaluation for Large Language Models
  8. 2025Multi-Value Alignment for LLMs via Value Decorrelation and Extrapolation