ECLAIR: Enhanced Clarification for Interactive Responses
This addresses ambiguity resolution for enterprise AI assistants, though it appears incremental as it builds on existing interactive disambiguation concepts with a generalized architecture.
The researchers tackled the problem of interactive disambiguation in enterprise AI assistants by developing ECLAIR, a unified framework that generates clarification questions for ambiguous queries and resolves ambiguity based on user responses, demonstrating superior performance compared to few-shot prompting techniques.
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.