SIMMER: Benchmarking Latent Failures in LLM Executable Planning with a World Model

Large language models (LLMs) are increasingly deployed as planners for autonomous agents in household environments. While existing benchmarks evaluate whether LLM-generated plans execute successfully, they overlook a critical type of failure: $\textit{latent failures}$. Unlike immediate failures that trigger instant feedback at execution time and enable timely correction, latent failures do not immediately halt plan execution but silently compromise goal achievement. In severe cases, they cause irreversible harm. To address this gap, we introduce $\textbf{SIMMER}$, a benchmark for evaluating latent failures in LLM planning through a human-curated symbolic world model grounded in the kitchen domain. \textsc{Simmer} defines a world model comprising 77 actions, 262 unique objects, and approximately 46,800 possible interactions that are semantically realistic, derived from real-world cooking scripts. It then leverages a state machine executor that validates plans against the world model and detects immediate precondition violations, latent hazards, and irreversible failures. Experiments across six LLMs show that even frontier models achieve at most 17\% error-free plans. Moreover, up to 56\% of plans contain latent failures, the majority of which lead to irreversible consequences. We further demonstrate that explicit state reasoning via counterfactual foresight simulation can reduce latent failures by up to 72\% and irreversible cases by up to 75\%, suggesting a promising direction for more robust LLM planners.