SymAgent: A Neural-Symbolic Self-Learning Agent Framework for Complex Reasoning over Knowledge Graphs

TL;DR

SymAgent is introduced, an innovative neural-symbolic agent framework that achieves collaborative augmentation between KGs and LLMs and conceptualize KGs as dynamic environments and transform complex reasoning tasks into a multi-step interactive process, enabling KGs to participate deeply in the reasoning process.

Abstract

Recent advancements have highlighted that Large Language Models (LLMs) are prone to hallucinations when solving complex reasoning problems, leading to erroneous results. To tackle this issue, researchers incorporate Knowledge Graphs (KGs) to improve the reasoning ability of LLMs. However, existing methods face two limitations: 1) they typically assume that all answers to the questions are contained in KGs, neglecting the incompleteness issue of KGs, and 2) they treat the KG as a static repository and overlook the implicit logical reasoning structures inherent in KGs. In this paper, we introduce SymAgent, an innovative neural-symbolic agent framework that achieves collaborative augmentation between KGs and LLMs. We conceptualize KGs as dynamic environments and transform complex reasoning tasks into a multi-step interactive process, enabling KGs to participate deeply in the reasoning process. SymAgent consists of two modules: Agent-Planner and Agent-Executor. The Agent-Planner leverages LLM's inductive reasoning capability to extract symbolic rules from KGs, guiding efficient question decomposition. The Agent-Executor autonomously invokes predefined action tools to integrate information from KGs and external documents, addressing the issues of KG incompleteness. Furthermore, we design a self-learning framework comprising online exploration and offline iterative policy updating phases, enabling the agent to automatically synthesize reasoning trajectories and improve performance. Experimental results demonstrate that SymAgent with weak LLM backbones (i.e., 7B series) yields better or comparable performance compared to various strong baselines. Further analysis reveals that our agent can identify missing triples, facilitating automatic KG updates.

Figures (7)

• Figure 1: Comparison between SymAgent and existing methods. Armed with an action tool library, the SymAgent, consisting of a planner and an executor, autonomously interacts with the KG environment to conduct reasoning.
• Figure 2: The overview of our proposed SymAgent. (a) the planner in SymAgent, which derives the symbolic rules from the KG to guide the reasoning; (b) the executor in SymAgent, which conducts the automatic action invocation to obtain the answer; (c) the self-learning framework to enhance the agent iteratively; and (d) an example of the synthesized action invoking trajectory.
• Figure 3: The impact of the iteration numbers in the self-learning phase on model performance.
• Figure 4: Performance of RoG on KG augmented with triples extracted by our model.
• Figure 5: Prompt template of our SymAgent.