# Value-Based Rationales Improve Social Experience: A Multiagent Simulation Study

arXiv: https://arxiv.org/abs/2408.02117 · HTML: https://mlsanity.com/paper/2408.02117/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 50 references that are in the map.
- First posted on arXiv 4 Aug 2024.

## Builds on (most cited first, 7 of 7)

- [TED: Teaching AI to Explain its Decisions](https://mlsanity.com/paper/1811.04896/) (2018)
- [Probabilistic Sufficient Explanations](https://mlsanity.com/paper/2105.10118/) (2021)
- [A computational framework for human values](https://mlsanity.com/paper/2305.02748/) (2023)
- [Macro Ethics Principles for Responsible AI Systems: Taxonomy and Future Directions](https://mlsanity.com/paper/2208.12616/) (2022)
- [Norm Enforcement with a Soft Touch: Faster Emergence, Happier Agents](https://mlsanity.com/paper/2401.16461/) (2024)
- [Socially Intelligent Genetic Agents for the Emergence of Explicit Norms](https://mlsanity.com/paper/2208.03789/) (2022)
- [PACCART: Reinforcing Trust in Multiuser Privacy Agreement Systems](https://mlsanity.com/paper/2302.13650/) (2023)

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

- [Rethinking How AI Embeds and Adapts to Human Values: Challenges and Opportunities](https://mlsanity.com/paper/2508.17104/) (2025)

## Abstract (arXiv)

We propose Exanna, a framework to realize agents that incorporate values in decision making. An Exannaagent considers the values of itself and others when providing rationales for its actions and evaluating the rationales provided by others. Via multiagent simulation, we demonstrate that considering values in decision making and producing rationales, especially for norm-deviating actions, leads to (1) higher conflict resolution, (2) better social experience, (3) higher privacy, and (4) higher flexibility.