SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914382429028352 |
|---|---|
| author | Fan, Jialiang Xu, Weizhe Liu, Mengyu Sokolsky, Oleg Lee, Insup Kong, Fanxin |
| author_facet | Fan, Jialiang Xu, Weizhe Liu, Mengyu Sokolsky, Oleg Lee, Insup Kong, Fanxin |
| contents | Safety-critical task planning in robotic systems remains challenging: classical planners suffer from poor scalability, Reinforcement Learning (RL)-based methods generalize poorly, and base Large Language Models (LLMs) cannot guarantee safety. To address this gap, we propose safety-generalizable large language models, named SafeGen-LLM. SafeGen-LLM can not only enhance the safety satisfaction of task plans but also generalize well to novel safety properties in various domains. We first construct a multi-domain Planning Domain Definition Language 3 (PDDL3) benchmark with explicit safety constraints. Then, we introduce a two-stage post-training framework: Supervised Fine-Tuning (SFT) on a constraint-compliant planning dataset to learn planning syntax and semantics, and Group Relative Policy Optimization (GRPO) guided by fine-grained reward machines derived from formal verification to enforce safety alignment and by curriculum learning to better handle complex tasks. Extensive experiments show that SafeGen-LLM achieves strong safety generalization and outperforms frontier proprietary baselines across multi-domain planning tasks and multiple input formats (e.g., PDDLs and natural language). |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_24235 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems Fan, Jialiang Xu, Weizhe Liu, Mengyu Sokolsky, Oleg Lee, Insup Kong, Fanxin Robotics Artificial Intelligence Safety-critical task planning in robotic systems remains challenging: classical planners suffer from poor scalability, Reinforcement Learning (RL)-based methods generalize poorly, and base Large Language Models (LLMs) cannot guarantee safety. To address this gap, we propose safety-generalizable large language models, named SafeGen-LLM. SafeGen-LLM can not only enhance the safety satisfaction of task plans but also generalize well to novel safety properties in various domains. We first construct a multi-domain Planning Domain Definition Language 3 (PDDL3) benchmark with explicit safety constraints. Then, we introduce a two-stage post-training framework: Supervised Fine-Tuning (SFT) on a constraint-compliant planning dataset to learn planning syntax and semantics, and Group Relative Policy Optimization (GRPO) guided by fine-grained reward machines derived from formal verification to enforce safety alignment and by curriculum learning to better handle complex tasks. Extensive experiments show that SafeGen-LLM achieves strong safety generalization and outperforms frontier proprietary baselines across multi-domain planning tasks and multiple input formats (e.g., PDDLs and natural language). |
| title | SafeGen-LLM: Enhancing Safety Generalization in Task Planning for Robotic Systems |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2602.24235 |