TreeMind: Automatically Reproducing Android Bug Reports via LLM-empowered Monte Carlo Tree Search
Zhengyu Chen, Zhaoyi Meng, Wenxiang Zhao, Wansen Wang, Wenchao Huang, Jie Cui, Hong Zhong, Yan Xiong
TL;DR
TreeMind tackles automatic reproduction of Android crashes from incomplete bug reports by marrying LLM-based semantic reasoning with an adapted Monte Carlo Tree Search. It introduces two LLM-guided agents, Expander and Simulator, to drive expansion and simulation within a goal-directed search over multi-modal UI states, enabling strategic planning where purely prompt-based methods struggle. Extensive experiments on 93 real-world bug reports show TreeMind achieving a reproduction success rate of 64.52%, outperforming four strong baselines, with detailed ablations confirming the value of each component. The work highlights the practical potential of integrating external planning with LLM reasoning to improve reliability and scalability in automated bug reproduction and debugging workflows.
Abstract
Automatically reproducing Android app crashes from textual bug reports is challenging, particularly when the reports are incomplete and the modern UI exhibits high combinatorial complexity. Existing approaches based solely on reinforcement learning or large language models (LLMs) exhibit limitations in such scenarios. They struggle to infer unobserved steps and reconstruct the underlying user action sequences to navigate the vast UI interaction space, primarily due to limited goal-directed reasoning and planning. We present TreeMind, a novel technique that integrates LLMs with an adapted Monte Carlo Tree Search (MCTS) algorithm to achieve strategic UI exploration in bug reproduction. To the best of our knowledge, this is the first work to combine external decision-making with LLM semantic reasoning for reliable and accurate reproduction processes. We formulate the reproduction task as a target-driven search problem, leveraging MCTS as the core planning mechanism to iteratively refine action sequences. To enhance MCTS with semantic reasoning, we introduce two LLM-guided agents with distinct roles: Expander generates top-k promising actions based on the current UI state and exploration history, while Simulator estimates the likelihood that each candidate action leads toward successful reproduction by additionally leveraging dynamic environment feedback. By incorporating multi-modal UI inputs and tailored prompting strategies, TreeMind performs feedback-aware navigation that identifies essential user actions and incrementally reconstructs reproduction paths. We evaluate TreeMind on a dataset of 93 real-world Android bug reports from three widely-used benchmarks. Experimental results show that it significantly outperforms four state-of-the-art baselines, including ReBL, ReActDroid, AdbGPT, and ReproBot, in reproduction success rate.