8th AAAI/ACM Conference on AI, Ethics, and Society

October 20-22, 2025 in Madrid, Spain

The 8th AAAI/ACM Conference on AI, Ethics, and Society (AIES) examines how artificial intelligence can be developed and deployed responsibly, with fairness, transparency, and alignment at the core. Kitware is pleased to have a paper accepted to this conference on steerable pluralism for LLM alignment. This work represents a natural extension of Kitware’s leadership in LLMs and Generative AI.

Visit our blog to learn more about our responsible AI and LLM projects.

Kitware’s Activities and Involvement

Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression

Conference Paper
Authors: Jadie Adams, Brian Hu, Emily Veenhuis, David Joy, Bharadwaj Ravichandran, Aaron Bray, Anthony Hoogs, Arslan Basharat

Traditional methods for aligning large language models (LLMs), such as reinforcement learning from human feedback (RLHF), typically optimize for average user preferences through scalar rewards. This limits the ability to represent the full spectrum of diverse values and priorities that real-world users bring to the table. Our paper proposes a pluralistic alignment approach that enables LLMs to adapt to multiple, sometimes competing, user perspectives.

The core of this work is a few-shot comparative regression method that allows models to reason across fine-grained attributes and steer responses accordingly. By leveraging in-context learning, we show how LLMs can be aligned not just for “helpfulness and harmlessness,” but for pluralism—capturing the nuances of diverse user preferences. To validate the approach, we introduce two new benchmarks, adapted from the Moral Integrity Corpus (MIC) and HelpSteer2, and demonstrate superior performance compared to state-of-the-art baselines.

Leaders in Artificial Intelligence

With more than 20 years of experience, Kitware is a leader in artificial intelligence. Our technical areas of focus include:

• AI Test, Evaluation, and Assurance
• Generative AI
• Interactive AI and Human-Machine Teaming
• Responsible and Explainable AI

Kitware’s AI expertise spans computer vision, machine learning, natural language processing, generative AI, medical and biomedical systems, scientific simulations, and software and data engineering.

Through our commitment to academic-level R&D and open source technologies, we ensure that our deployed solutions remain at the cutting edge, empowering our partners to achieve their goals with the most advanced AI capabilities.

Visit our blog to learn more about our work in this area.