Answering Questions over Knowledge Graphs (KGQA) is key to well-functioning
autonomous language agents in various real-life applications. To improve the
neural-symbolic reasoning capabilities of language agents powered by Large
Language Models (LLMs) in KGQA, we propose the DecompositionAlignment-Reasoning
Agent (DARA) framework. DARA effectively parses questions into formal queries
through a dual mechanism: high-level iterative task decomposition and low-level
task grounding. Importantly, DARA can be efficiently trained with a small
number of high-quality reasoning trajectories. Our experimental results
demonstrate that DARA fine-tuned on LLMs (e.g. Llama-2-7B, Mistral) outperforms
both in-context learning-based agents with GPT-4 and alternative fine-tuned
agents, across different benchmarks in zero-shot evaluation, making such models
more accessible for real-life applications. We also show that DARA attains
performance comparable to state-of-the-art enumerating-and-ranking-based
methods for KGQA.

回答知识图谱上的问题（KGQA）是不同实际应用中实现良好性能的语言智能代理的关键。为了提高由大型语言模型（LLMs）驱动的语言智能代理在 KGQA 中的神经符号推理能力，我们提出了 DecompositionAlignment-Reasoning Agent（DARA）框架。DARA 通过高级迭代任务分解和低级任务基础形式将问题有效地解析为正式查询。重要的是，DARA 可以通过少量高质量的推理轨迹进行高效训练。我们的实验结果表明，DARA 在 LLMs 上进行微调（例如 Llama-2-7B，Mistral）在零 - shot 评估的不同基准中优于基于上下文学习的代理（如 GPT-4）和其他微调代理，从而使这样的模型在实际应用中更易访问。我们还显示出，DARA 在 KGQA 方面达到了与最先进的枚举和排名方法相当的性能。