# HAPI: A Model for Learning Robot Facial Expressions from Human Preferences

arXiv: https://arxiv.org/abs/2503.17046 · HTML: https://mlsanity.com/paper/2503.17046/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 29 references that are in the map.
- First posted on arXiv 21 Mar 2025.

## Builds on (most cited first, 8 of 8)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Deep Facial Expression Recognition: A Survey](https://mlsanity.com/paper/1804.08348/) (2018)
- [Primitive Skill-based Robot Learning from Human Evaluative Feedback](https://mlsanity.com/paper/2307.15801/) (2023)
- [Autonomous Robotic Reinforcement Learning with Asynchronous Human Feedback](https://mlsanity.com/paper/2310.20608/) (2023)
- [Smile Like You Mean It: Driving Animatronic Robotic Face with Learned Models](https://mlsanity.com/paper/2105.12724/) (2021)
- [Optimizing Facial Expressions of an Android Robot Effectively: a Bayesian Optimization Approach](https://mlsanity.com/paper/2301.05620/) (2023)

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

- [Eye, Robot: Learning Expressive Gaze Dynamics via a VLM-Guided Transformer-VAE](https://mlsanity.com/paper/2610.19635/) (2026)

## Abstract (arXiv)

Automatic robotic facial expression generation is crucial for human-robot interaction, as handcrafted methods based on fixed joint configurations often yield rigid and unnatural behaviors. Although recent automated techniques reduce the need for manual tuning, they tend to fall short by not adequately bridging the gap between human preferences and model predictions-resulting in a deficiency of nuanced and realistic expressions due to limited degrees of freedom and insufficient perceptual integration. In this work, we propose a novel learning-to-rank framework that leverages human feedback to address this discrepancy and enhanced the expressiveness of robotic faces. Specifically, we conduct pairwise comparison annotations to collect human preference data and develop the Human Affective Pairwise Impressions (HAPI) model, a Siamese RankNet-based approach that refines expression evaluation. Results obtained via Bayesian Optimization and online expression survey on a 35-DOF android platform demonstrate that our approach produces significantly more realistic and socially resonant expressions of Anger, Happiness, and Surprise than those generated by baseline and expert-designed methods. This confirms that our framework effectively bridges the gap between human preferences and model predictions while robustly aligning robotic expression generation with human affective responses.