# Democratic ICAI: Debating Our Way to Steering Principles from Preferences

arXiv: https://arxiv.org/abs/2606.28294 · HTML: https://mlsanity.com/paper/2606.28294/

## 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 27 of its 47 references that are in the map.
- First posted on arXiv 26 Jun 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, 27 of 27)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Learning to summarize from human feedback](https://mlsanity.com/paper/2009.01325/) (2020)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Towards Understanding Sycophancy in Language Models](https://mlsanity.com/paper/2310.13548/) (2023)
- [Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate](https://mlsanity.com/paper/2305.19118/) (2023)
- [Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluators](https://mlsanity.com/paper/2404.04475/) (2024)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [AI safety via debate](https://mlsanity.com/paper/1805.00899/) (2018)
- [Reward Shaping to Mitigate Reward Hacking in RLHF](https://mlsanity.com/paper/2502.18770/) (2025)
- [RRM: Robust Reward Model Training Mitigates Reward Hacking](https://mlsanity.com/paper/2409.13156/) (2024)
- [Can You Trust LLM Judgments? Reliability of LLM-as-a-Judge](https://mlsanity.com/paper/2412.12509/) (2024)
- [LitBench: A Benchmark and Dataset for Reliable Evaluation of Creative Writing](https://mlsanity.com/paper/2507.00769/) (2025)
- [Inverse Constitutional AI: Compressing Preferences into Principles](https://mlsanity.com/paper/2406.06560/) (2024)
- [Auto-Rubric: Learning From Implicit Weights to Explicit Rubrics for Reward Modeling](https://mlsanity.com/paper/2510.17314/) (2025)
- [The Need for a Socially-Grounded Persona Framework for User Simulation](https://mlsanity.com/paper/2601.07110/) (2026)
- [GPTree: Towards Explainable Decision-Making via LLM-powered Decision Trees](https://mlsanity.com/paper/2411.08257/) (2024)
- [Improving Language Model Personas via Rationalization with Psychological Scaffolds](https://mlsanity.com/paper/2504.17993/) (2025)
- [Decoding Human Preferences in Alignment: An Improved Approach to Inverse Constitutional AI](https://mlsanity.com/paper/2501.17112/) (2025)
- [Beyond Preferences: Learning Alignment Principles Grounded in Human Reasons and Values](https://mlsanity.com/paper/2601.18760/) (2026)

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

- [Open Problems in Constitutional Preference Reconstruction](https://mlsanity.com/paper/2606.30116/) (2026)

## Abstract (arXiv)

Preference-based alignment often struggles to capture the reasoning that underlies human judgments. Many evaluations rely on multiple interacting criteria, yet pairwise labels reveal only the final choice rather than the considerations that shape preferences. Inverse Constitutional AI (ICAI) improves interpretability in decision making by summarizing preferences into natural-language principles, but its single-pass explanations miss much of the nuance involved in complex decisions. We introduce Democratic ICAI (DICAI), a novel approach that gathers multiple competing rationales through structured persona debate, offering a broader and more expressive account of the factors influencing each comparison. From these richer signals, we derive clearer and more comprehensive steering principles and use them to guide preference decision modeling through both LLM-based and decision-tree judges, as well as downstream model training via constitution-induced preference labels. Experiments on creative preference benchmarks, MuCE-Pref and LiTBench, across multiple creative task categories show that Democratic ICAI yields a more faithful preference structure. It improves average preference prediction across tasks relative to deliberative prompting and principle-based baselines, while producing constitutions that LLM annotators prefer.