Towards AI Search Paradigm
🤖 AI Summary
Existing AI search systems struggle to simultaneously support both simple queries and complex, multi-step reasoning tasks. Method: This paper proposes a human-cognitive-inspired next-generation search paradigm, implemented via a modular LLM agent architecture comprising four coordinated roles—Master, Planner, Executor, and Writer—enabling query complexity awareness, automatic task decomposition, dynamic tool orchestration, and robust RAG optimization. Contribution/Results: We formally define the first design principles for trustworthy, adaptive, and scalable AI search systems; establish a systematic methodology for dynamic workflow execution; and demonstrate significant improvements over conventional single-model or static RAG architectures in accuracy, task generalization, and decision interpretability. Extensive experiments validate effectiveness across multi-stage reasoning and real-world decision-making scenarios.
Application Category
📝 Abstract
In this paper, we introduce the AI Search Paradigm, a comprehensive blueprint for next-generation search systems capable of emulating human information processing and decision-making. The paradigm employs a modular architecture of four LLM-powered agents (Master, Planner, Executor and Writer) that dynamically adapt to the full spectrum of information needs, from simple factual queries to complex multi-stage reasoning tasks. These agents collaborate dynamically through coordinated workflows to evaluate query complexity, decompose problems into executable plans, and orchestrate tool usage, task execution, and content synthesis. We systematically present key methodologies for realizing this paradigm, including task planning and tool integration, execution strategies, aligned and robust retrieval-augmented generation, and efficient LLM inference, spanning both algorithmic techniques and infrastructure-level optimizations. By providing an in-depth guide to these foundational components, this work aims to inform the development of trustworthy, adaptive, and scalable AI search systems.
Problem
Research questions and friction points this paper is trying to address.
Develop next-gen search systems emulating human cognition
Modular LLM agents handle diverse information needs
Integrate planning, execution, and synthesis for complex queries
Innovation
Methods, ideas, or system contributions that make the work stand out.
Modular LLM-powered agents for dynamic adaptation
Coordinated workflows for query evaluation and execution
Algorithmic and infrastructure-level optimizations for efficiency
🔎 Similar Papers
No similar papers found.
