ATLAS: Actor-Critic Task-Completion with Look-ahead Action Simulation

ICLR 2026 Conference Submission22460 Authors

Keywords: LLM Agent, Planning, Tree Search, Cognitive Map

Abstract: We observe that current state-of-the-art web-agents are unable to effectively adapt to new environments without neural network fine-tuning, without which they produce inefficient execution plans due to a lack of awareness of the structure and dynamics of the new environment. To address this limitation, we introduce \textbf{ATLAS} (\textbf{A}ctor-Critic \textbf{T}ask-completion with \textbf{L}ook-ahead \textbf{A}ction \textbf{S}imulation), a memory-augmented agent that is able to make plans grounded in a \emph{model} of the environment by simulating the consequences of those actions in \emph{cognitive space}. Our agent starts by building a \emph{"cognitive map"} by performing a lightweight curiosity driven exploration of the environment. The planner proposes candidate actions; the simulator predicts their consequences in cognitive space; a critic analyzes the options to select the best roll-out and update the original plan; and a browser executor performs the chosen action. On the \textbf{WebArena-Lite} Benchmark, we achieve a 63\% success rate compared to ~53.9\% success rate for the previously published state-of-the-art. Unlike previous systems, our modular architecture requires no website-specific LLM fine-tuning. Ablations show sizable drops without the world-model, hierarchical planner, and look-ahead-based replanner confirming their complementary roles within the design of our system.

Primary Area: applications to robotics, autonomy, planning

Submission Number: 22460