Read about a method for eliciting a robot's personality through a short conversation rather than a survey.
Understand how a structured PersonaSpec captures traits, values, and behavioral policies for a robot.
Reference PACE's approach when designing adaptive, emotionally expressive robot personas.
| lipzh5/pace | 0xdea/ttyinject-rs | 0xhossam/uncanny | |
|---|---|---|---|
| Stars | 12 | 12 | 12 |
| Language | — | Rust | C |
| Setup difficulty | hard | moderate | hard |
| Complexity | 4/5 | 4/5 | 5/5 |
| Audience | researcher | researcher | researcher |
Figures from each repo's GitHub metadata at analysis time.
No installation steps or runnable code are included, this describes a research method deployed on the Ameca humanoid robot.
PACE is a research project about giving humanoid robots a personality that adapts to the person they are talking to, instead of using one fixed personality written in advance by a developer. The repository accompanies an academic paper called PACE: Persona Adaptation through Conversational Elicitation in Human Robot Interaction, and the README describes the method rather than providing a ready to run application. The approach works in three stages. First, the robot asks a short set of open ended questions, then generates follow up questions if it needs more detail, rather than making the person fill out a long personality survey. Second, the answers from that conversation are turned into a structured description called a PersonaSpec, which captures psychological and behavioral details such as personality traits, values, motivations, how the person tends to regulate their behavior, how they see their own identity, and rules for how to act in different situations. Third, that structured persona description is compiled into a system prompt, the kind of instruction text used to steer a large language model, and the robot then speaks and expresses facial expressions that match the generated persona. The README lists supporting details about the system: it is designed to run on the Ameca humanoid robot, it uses Amazon Polly for generating speech, it uses a language model called GPT-5.5-mini for generating the persona specification, and it selects facial expressions from seven basic emotion categories, choosing them based on the context of what is being said so that speech and facial animation stay in sync. A project page with a video demo is linked, hosted anonymously, which is common practice for research submitted for anonymous peer review. The README does not show installation steps, code files, or a license, and reads as a description of the research method and system design rather than documentation for running the code yourself. It would mainly interest researchers and engineers working on robotics, human-robot interaction, or personalized AI agents.
A research project describing PACE, a method for giving humanoid robots an adaptive persona built from a short conversation instead of a fixed system prompt.
No license information is stated in the README.
Setup difficulty is rated hard, with roughly 1day+ to a first successful run.
Mainly researcher.
This repo across BitVibe Labs
Don't trust strangers blindly. Verify against the repo.