Skip to content

PlanGuard: A Guardrail for Multi-Step Plan Safety in Embodied Agents

Sep 2026 · 0 citations · 24 references
Computer Science

TL;DR

This work introduces PlanGuard, the first pre-execution detector that evaluates the physical safety of a complete multi-step plan in its current environment, and proposes Strong-Teacher Adaptive Compensation for On-Policy Distillation (STAC-OPD), which provides compact models with adaptive strong-teacher supervision along their on-policy trajectories.

Abstract

Embodied task planners may produce multi-step plans whose subtask dependencies and interactions with the environment create physical risks during execution. Yet existing safeguards overlook such compositional risks, as general-purpose guardrails focus on semantic harm and embodied safety detectors assess subtasks in isolation. To address this gap, we introduce PlanGuard, the first pre-execution detector that evaluates the physical safety of a complete multi-step plan in its current environment. For training and evaluation, we construct a Multi-Step Plan Safety (MSP-Safe) dataset through paired task construction, plan generation using diverse planners, and safety annotation by three judges. Task-oriented SFT on MSP-Safe establishes fundamental plan-safety assessment capabilities, yet a substantial gap remains between compact models suitable for real-time deployment and stronger but costlier large models. Accordingly, we propose Strong-Teacher Adaptive Compensation for On-Policy Distillation (STAC-OPD), which provides compact models with adaptive strong-teacher supervision along their on-policy trajectories. It combines token-level distribution transfer from a fine-tuned strong teacher with probability-routed sequence-level compensation, retaining student-generated targets when the student favors the reference safety decision and using teacher-reconstructed targets otherwise. Across all test subsets, PlanGuard-2B achieves average 87.15% ACC and 87.21% F1, demonstrating effective whole-plan physical-risk detection at compact model scale. Code and dataset will be publicly released.

View source

Similar papers

#machine learning Preprint Oct 2026

Mind the Refinement Gap: When Safe High-Level Robot Plans Produce Unsafe Executions

Language-enabled robot systems increasingly combine semantic-graph planning with temporal-logic safety monitors. We investigate a trace-completeness assumption in these systems: whether the high-level action sequence checked by a monitor represents the navigation and implicit action effects induced during execution. We...

Stabak Das, Priyesh Ranjan, Xiang-Fang Li et al. · 0 citations
#computer vision Preprint Aug 2026

SafeBranch: Branch-Pair Safety Alignment for Embodied Agents

SafeBranch is proposed, a framework that aligns an embodied actor on safety through branch pairs constructed from the actor's own unsafe rollouts via environment rollback, achieving roughly ten times more safe successes than the untrained baseline on the unseen-object variant.

Hyunse Lee, Jiwoo Jeong, Haneul Lee et al. · 0 citations
Preprint Aug 2026

MANIGUARD: A Benchmark and Data Suite for Specification-Grounded Safety Evaluation and Improvement of Robotic Manipulation

Foundation-model policies for robotic manipulation are advancing rapidly on task success, but rigorous evaluation of whether they succeed safely is still lacking. We introduce ManiGuard, a specification-grounded framework for evaluating and improving the safety of foundation-model manipulation, comprising the ManiGuard...

Yi Peng, Philip Wang, Simon Sinong Zhan et al. · 2 citations · ⚡1

SAGE: Symbolic Action-Gating and Editing for LLM Task Planners

SAGE (Symbolic Action-Gating and Editing), a single-LLM planner built from two lightweight mechanisms: a domain-agnostic symbolic gate that blocks precondition-violating actions with typed reasons as a runtime safety monitor, and a local edit that regenerates only the failed sub-goal's suffix, keeping completed and unt...

T. Bui, Jongsul Moon, Youngouk Kim et al. · 0 citations
Preprint Sep 2026

Safe Task Planning with Long-Term Graph Memory for Embodied Agents

A novel safe task-planning framework, SafeMem, which constructs and maintains a long-term semantic graph memory of the open and dynamic environment, and substantially improves safe success rates compared to state-of-the-art VLM-driven task planners.

Si-Yuan Li, Tai-Yan Lang, Ao Yan et al. · 1 citation

AgentBoundary: Counterfactual Evaluation of Safety in Tool-Using LLM Agents

This work introduces AgentBound, the first four-way counterfactual generation-and-evaluation framework for tool-using agent safety, and shows that effective agentic alignment requires action decisions to track permission-relevant execution evidence, rather than refusal strength alone.

Tian-Zhuo Yang, Zi-Rui Mi, Yan-Tao Huang et al. · 1 citation

Related blog posts

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.