DeepSolution: Boosting Complex Engineering Solution Design via Tree-based Exploration and Bi-point Thinking

2025-02-28

AI Summary

To address the challenge that conventional RAG methods struggle to generate complete and feasible solutions for multi-constrained complex engineering problems, this paper proposes SolutionRAG—a novel retrieval-augmented generation system. Methodologically, it introduces a dual-point thinking–driven tree-based exploration architecture that jointly models feasibility and innovativeness; constructs SolutionBench, the first end-to-end evaluation benchmark specifically designed for engineering solution generation; and designs a constraint-aware hierarchical reasoning mechanism with completeness guarantees. Evaluated on SolutionBench, SolutionRAG significantly outperforms state-of-the-art RAG baselines and fine-tuned large language models: solution completeness improves by 37.2%, feasibility by 41.5%, achieving new state-of-the-art performance.

Technology Category

• Constraint Satisfaction and Optimization: Solvers and Tools
• Search and Optimization: Metareasoning and Metaheuristics
• Knowledge Representation and Reasoning: Knowledge Engineering

Application Category

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

Abstract

Designing solutions for complex engineering challenges is crucial in human production activities. However, previous research in the retrieval-augmented generation (RAG) field has not sufficiently addressed tasks related to the design of complex engineering solutions. To fill this gap, we introduce a new benchmark, SolutionBench, to evaluate a system's ability to generate complete and feasible solutions for engineering problems with multiple complex constraints. To further advance the design of complex engineering solutions, we propose a novel system, SolutionRAG, that leverages the tree-based exploration and bi-point thinking mechanism to generate reliable solutions. Extensive experimental results demonstrate that SolutionRAG achieves state-of-the-art (SOTA) performance on the SolutionBench, highlighting its potential to enhance the automation and reliability of complex engineering solution design in real-world applications.

Problem

• Addresses complex engineering solution design challenges
• Introduces SolutionBench for evaluating solution generation
• Proposes SolutionRAG for reliable engineering solutions

Innovation

• Introduces SolutionBench for engineering solution evaluation
• Develops SolutionRAG with tree-based exploration mechanism
• Employs bi-point thinking for reliable solution generation