Level-Navi Agent: A Framework and benchmark for Chinese Web Search Agents

TL;DR

This paper introduces Level-Navi, a framework and benchmark for Chinese web search agents using LLMs, including a new dataset and evaluation metric to improve understanding and performance.

Contribution

It presents a training-free, level-aware web search agent framework, a well-annotated dataset (Web24), and a new evaluation metric for Chinese web search agents.

Findings

Level-Navi effectively handles complex queries across internet levels.

State-of-the-art LLMs are comprehensively evaluated under fair settings.

The dataset and metric facilitate future research in Chinese web search agents.

Abstract

Large language models (LLMs), adopted to understand human language, drive the development of artificial intelligence (AI) web search agents. Compared to traditional search engines, LLM-powered AI search agents are capable of understanding and responding to complex queries with greater depth, enabling more accurate operations and better context recognition. However, little attention and effort has been paid to the Chinese web search, which results in that the capabilities of open-source models have not been uniformly and fairly evaluated. The difficulty lies in lacking three aspects: an unified agent framework, an accurately labeled dataset, and a suitable evaluation metric. To address these issues, we propose a general-purpose and training-free web search agent by level-aware navigation, Level-Navi Agent, accompanied by a well-annotated dataset (Web24) and a suitable evaluation metric. Level-Navi Agent can think through complex user questions and conduct searches across various levels on the internet to gather information for questions. Meanwhile, we provide a comprehensive evaluation of state-of-the-art LLMs under fair settings. To further facilitate future research, source code is available at Github.