ECLAIR: Enhanced Clarification for Interactive Responses

2025-03-19

AI Summary

This paper addresses the problem of ambiguous user queries degrading response accuracy in enterprise AI assistants. We propose the first end-to-end interactive disambiguation framework. Methodologically, it employs a large language model–based joint generation-and-reasoning architecture, integrating multi-agent collaborative information, domain-agent-driven semantic grounding, and ambiguity-aware contextual modeling; it further enables dynamic clarification via end-to-end fine-tuning and interactive reinforcement learning. Our key contribution is a unified modeling of clarification generation, user feedback interpretation, and ambiguity resolution—supporting customizable domain knowledge injection and seamless integration with multiple downstream agents. Experiments demonstrate significant improvements over few-shot prompting baselines on both clarification question generation and ambiguity resolution tasks, achieving higher accuracy and interaction efficiency. The framework’s effectiveness is validated on real-world enterprise dialogue data.

Technology Category

• Natural Language Processing: Language Grounding & Multi-modal NLP
• Humans and AI: Intelligent User Interfaces
• Data Mining & Knowledge Management: Intelligent Query Processing

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We present ECLAIR (Enhanced CLArification for Interactive Responses), a novel unified and end-to-end framework for interactive disambiguation in enterprise AI assistants. ECLAIR generates clarification questions for ambiguous user queries and resolves ambiguity based on the user's response.We introduce a generalized architecture capable of integrating ambiguity information from multiple downstream agents, enhancing context-awareness in resolving ambiguities and allowing enterprise specific definition of agents. We further define agents within our system that provide domain-specific grounding information. We conduct experiments comparing ECLAIR to few-shot prompting techniques and demonstrate ECLAIR's superior performance in clarification question generation and ambiguity resolution.

Problem

• Develops a framework for interactive disambiguation in AI assistants.
• Generates clarification questions for ambiguous user queries.
• Integrates ambiguity information from multiple domain-specific agents.

Innovation

• Unified end-to-end framework for disambiguation
• Generates clarification questions for ambiguous queries
• Integrates ambiguity information from multiple agents