# Mitigating Response Delays in Free-Form Conversations with LLM-powered Intelligent Virtual Agents

**Authors:** Mykola Maslych, Mohammadreza Katebi, Christopher Lee, Yahya Hmaiti, Amirpouya Ghasemaghaei, Christian Pumarada, Janneese Palmer, Esteban Segarra Martinez, Marco Emporio, Warren Snipes, Ryan P. McMahan, Joseph J. LaViola  
**arXiv:** [2507.22352](https://arxiv.org/abs/2507.22352)  
**Date:** 2025-07-30  
**Summary page:** https://www.summarizepaper.com/en/paper/2507.22352v1/

## Key points

- Study focused on challenges of mitigating response delays in free-form conversations with virtual agents powered by Large Language Models (LLMs) within Virtual Reality (VR
- Researchers used conversational fillers such as gestures and verbal cues to bridge delays between user input and system responses
- Latency exceeding 4 seconds significantly degrades user experience quality
- Natural conversational fillers enhance perceived response time, especially in high-delay conditions
- Insights for practitioners and researchers seeking to optimize user engagement when conversational systems face delays due to network limitations or slow hardware
- Research team contributed an open-source pipeline to streamline deployment of conversational agents in virtual environments

## Summary

A study conducted by Mykola Maslych et al. investigated the challenges of mitigating response delays in free-form conversations with virtual agents powered by Large Language Models (LLMs) within Virtual Reality (VR). The researchers utilized conversational fillers such as gestures and verbal cues to bridge delays between user input and system responses. They evaluated their effectiveness across various latency levels and interaction scenarios, revealing that latency exceeding 4 seconds significantly degrades the quality of user experience. However, natural conversational fillers were found to enhance perceived response time, particularly in high-delay conditions. These findings offer valuable insights for practitioners and researchers seeking to optimize user engagement when conversational systems experience delays due to network limitations or slow hardware. Furthermore, the research team contributed an open-source pipeline designed to streamline the deployment of conversational agents in virtual environments. Their work was published at the 7th ACM Conference on Conversational User Interfaces (CUI '25) and is available for reference with DOI: 10.1145/3719160.3736636. This study sheds light on effective strategies for enhancing user interactions with intelligent virtual agents in immersive virtual reality settings.