Chat Failures and Troubles: Reasons and Solutions

TL;DR

The paper analyzes sociolinguistic and design-driven sources of chat failures in Human-Robot Interaction (HRI) and surveys limitations of traditional rule-based, generative, and large language model approaches. It highlights ten core problems, illustrated by well-known failures from Tay to ChatGPT, and argues for a unified mitigation framework. The proposed solution is a closed-loop control system with vocabulary filtering, online learning from data streams, continuous fine-tuning, and reinforcement learning to adaptively reduce errors while managing safety and autonomy. The work emphasizes continuous data-driven improvement and ontology-guided social vocabulary to enable robust, ethical, and context-aware HRI chat.

Abstract

This paper examines some common problems in Human-Robot Interaction (HRI) causing failures and troubles in Chat. A given use case's design decisions start with the suitable robot, the suitable chatting model, identifying common problems that cause failures, identifying potential solutions, and planning continuous improvement. In conclusion, it is recommended to use a closed-loop control algorithm that guides the use of trained Artificial Intelligence (AI) pre-trained models and provides vocabulary filtering, re-train batched models on new datasets, learn online from data streams, and/or use reinforcement learning models to self-update the trained models and reduce errors.