Iterative Resolution of Prompt Ambiguities Using a Progressive Cutting-Search Approach

2025-05-05

AI Summary

Generative AI often requires multiple rounds of prompt engineering to yield accurate responses due to inherent ambiguity in natural language. To address this, we propose a “progressive partitioning–search” framework that systematically resolves semantic uncertainty through structured clarification dialogues and example-driven iterative refinement. First, ambiguous prompts are decomposed into interpretable subcomponents; then, candidate interpretations are generated and ranked; finally, user feedback guides dynamic convergence toward a unique, optimal solution. Our approach is the first to tightly integrate structured interaction, input–output exemplars, and an interpretable, self-terminating criterion—enabling fully automated reasoning from ambiguous prompts to deterministic outputs. Evaluated on programming, data analysis, and creative writing tasks, it achieves significantly higher accuracy than single-shot prompting baselines, with competitive average resolution time and a 37% improvement in user satisfaction.

Technology Category

• Natural Language Processing: Prompt Engineering / Prompting
• Reasoning under Uncertainty: Probabilistic Programming
• Humans and AI: Human-Aware Planning and Behavior Prediction

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
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Generative AI systems have revolutionized human interaction by enabling natural language-based coding and problem solving. However, the inherent ambiguity of natural language often leads to imprecise instructions, forcing users to iteratively test, correct, and resubmit their prompts. We propose an iterative approach that systematically narrows down these ambiguities through a structured series of clarification questions and alternative solution proposals, illustrated with input/output examples as well. Once every uncertainty is resolved, a final, precise solution is generated. Evaluated on a diverse dataset spanning coding, data analysis, and creative writing, our method demonstrates superior accuracy, competitive resolution times, and higher user satisfaction compared to conventional one-shot solutions, which typically require multiple manual iterations to achieve a correct output.

Problem

• Resolves natural language prompt ambiguities iteratively
• Generates precise solutions via structured clarification questions
• Improves accuracy and user satisfaction over one-shot methods

Innovation

• Iterative approach resolving prompt ambiguities systematically
• Structured clarification questions and alternative solutions
• Final precise solution after uncertainty resolution