Agent Alpha: Unifying Generation, Exploration and Evaluation via Tree Search for Computer-Use Solution Discovery
Sizhe Tang · Rongqian chen · Tian Lan
Agent Alpha unifies generation, exploration, and evaluation via MCTS for computer-use solution discovery, enabling error recovery and prefix reuse through tree-informed reflection and comparison-driven evaluation, achieving ~77% on OSWorld.
摘要
Discovering solutions for complex computer-use tasks remains a fundamental challenge due to vast action spaces, non-reversible state transitions, and sparse rewards. Existing approaches for Graphical User Interface (GUI) agents operate as unidirectional processes, where the lack of regressive ability prevents the reuse of partial successes and recovery from early missteps. We introduce Agent Alpha, a unified framework that synthesis generation, exploration, and evaluation through step-level Monte Carlo Tree Search (MCTS) for efficient solution discovery. By integrating Alpha-UCT guided search into the interaction loop, Agent Alpha enables deliberate planning with early pruning of suboptimal branches and efficient prefix reuse. We further employ comparison-driven evaluation to mitigate absolute scoring biases and diversity-constrained expansion to maintain a compact, informative search space. We analyze the regret bound of Alpha-UCT and show it achieves tighter confidence bounds that accelerate identification of successful trajectories. On the OSWorld benchmark, Agent Alpha achieves a state-of-the-art success rate of ∼77%, significantly outperforming trajectory- level baselines. Code is available at: github.com/sizhe-tang/AgentAlpha.