SafeDec: Constrained Decoding for Safe Autoregressive Generalist Robot Policies

ICLR 2026 Conference Submission20629 Authors

Keywords: Generalist Robot Policies, Provably Safe Robotics, Neuro-symbolic Robotics

TL;DR: A constrained decoding framework for enforcing safety rules on transformer based robotic policies

Abstract:

Recent advances in end-to-end, multi-task robot policies based on transformer models have demonstrated impressive generalization to real-world embodied AI tasks. Trained on vast datasets of simulated and real-world trajectories, these models map multimodal observations directly to action sequences for physical execution. Despite promising real-world capabilities, these models are still data-driven and, therefore, lack explicit notions of behavioral correctness. We address this gap by introducing \textbf{SafeDec}, a constrained decoding framework for autoregressive, transformer-based robot policies that enforces invariant safety specifications on candidate action trajectories. Task-specific safety rules are expressed as Signal Temporal Logic (STL) formulas and are enforced at inference time with minimal overhead. Our method ensures that generated actions provably satisfy STL specifications under assumed dynamics at runtime without retraining , while remaining agnostic of the underlying policy.
We evaluate \textbf{SafeDec} on tasks from the CHORES benchmark for state-of-the-art generalist policies (e.g., SPOC, Flare, PoliFormer) across hundreds of procedurally generated environments and show that our decoding-time interventions are useful not only for filtering unsafe actions but also for conditional action generation. Videos are available at constrained-robot-fms.github.io.

Primary Area: applications to robotics, autonomy, planning

Submission Number: 20629