LogiDebrief: A Signal-Temporal Logic based Automated Debriefing Approach with Large Language Models Integration

Zirong Chen, Ziyan An, Jennifer Reynolds, Kristin Mullen, Stephen Martini, Meiyi Ma·May 06, 2025

Summary

LogiDebrief, an AI framework, automates 9-1-1 call debriefing with Signal-Temporal Logic & Large Language Models. Deployed at Metro Nashville, it saved 311.85 hours, outperforming traditional QA in user studies. It formalizes call-taking, identifies responder types, incident classifications, & critical conditions, ensuring compliance with STL-based checks. Future updates will incorporate human feedback for rule refinement.

Introduction
Background
Overview of 9-1-1 call handling processes
Challenges in traditional call debriefing methods
Objective
To present LogiDebrief, an AI framework that automates 9-1-1 call debriefing using Signal-Temporal Logic and Large Language Models
Method
Data Collection
Gathering 9-1-1 call data for analysis
Data Preprocessing
Cleaning and structuring the call data for AI processing
Model Development
Training Large Language Models for understanding call content
Incorporating Signal-Temporal Logic for formalizing call-taking processes
Deployment
Implementation at Metro Nashville
Performance metrics and outcomes
Results
Efficiency Improvements
Time savings achieved (311.85 hours)
Performance Comparison
Outperformance of traditional Quality Assurance methods
Call Debriefing Formalization
Identification of responder types, incident classifications, and critical conditions
Compliance Checks
Ensuring adherence to STL-based criteria
Future Enhancements
Human Feedback Integration
Plans for incorporating user feedback for rule refinement
Scalability and Adaptability
Strategies for expanding the framework's capabilities
Research and Development
Ongoing studies on AI in emergency call handling
Conclusion
Summary of LogiDebrief's Impact
Recap of LogiDebrief's benefits and achievements
Future Directions
Outlook on the evolution of AI in 9-1-1 call management
Basic info
papers
artificial intelligence
Advanced features
Insights
In what ways does LogiDebrief ensure compliance with STL-based checks during 9-1-1 call debriefing?
How does LogiDebrief utilize Signal-Temporal Logic and Large Language Models in its framework?
What are the key implementation features of LogiDebrief that contribute to its efficiency in automating 9-1-1 call debriefing?
How will future updates to LogiDebrief incorporate human feedback for rule refinement?