DR-Arena: an Automated Evaluation Framework for Deep Research Agents

TL;DR

DR-Arena is an automated, real-time evaluation framework for Deep Research agents that dynamically tests their reasoning and coverage capabilities using live web data, outperforming static benchmarks.

Contribution

It introduces a novel dynamic evaluation framework with real-time information trees and adaptive task escalation, addressing limitations of static datasets.

Findings

Achieves a Spearman correlation of 0.94 with human preferences.

Demonstrates reliable evaluation of DR agents without manual effort.

Validates DR-Arena as a state-of-the-art benchmarking tool.

Abstract

As Large Language Models (LLMs) increasingly operate as Deep Research (DR) Agents capable of autonomous investigation and information synthesis, reliable evaluation of their task performance has become a critical bottleneck. Current benchmarks predominantly rely on static datasets, which suffer from several limitations: limited task generality, temporal misalignment, and data contamination. To address these, we introduce DR-Arena, a fully automated evaluation framework that pushes DR agents to their capability limits through dynamic investigation. DR-Arena constructs real-time Information Trees from fresh web trends to ensure the evaluation rubric is synchronized with the live world state, and employs an automated Examiner to generate structured tasks testing two orthogonal capabilities: Deep reasoning and Wide coverage. DR-Arena further adopts Adaptive Evolvement Loop, a state-machine controller that dynamically escalates task complexity based on real-time performance, demanding deeper deduction or wider aggregation until a decisive capability boundary emerges. Experiments with six advanced DR agents demonstrate that DR-Arena achieves a Spearman correlation of 0.94 with the LMSYS Search Arena leaderboard. This represents the state-of-the-art alignment with human preferences without any manual efforts, validating DR-Arena as a reliable alternative for costly human adjudication.