Research map / Reinforcement learning / Visual and representation RL
Safe Reinforcement Learning with Natural Language Constraints
Tsung-Yen Yang, Michael Hu, Yinlam Chow, Peter Ramadge, Karthik R Narasimhan
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| remove the constraint mask M_C (ablation) | POLCO · HazardWorld-grid | constraint violation Delta_C: 66.67\% higher Delta_C without M_C | Improved |
| use full model (M_C, M_B, h_C embedding) versus ablated variants (ablation) | POLCO · HazardWorld-grid | reward and constraint violations: 5.12\% more reward and 2.22\% fewer constraint violations | Improved |
| learned constraint interpreter plus policy versus end-to-end constraint fusion with PCPO | POLCO · HazardWorld | reward and constraint violations: up to 11x reward, 1.8x fewer violations | Improved |
Closest papers by meaning
- From Text to Trajectory: Exploring Complex Constraint Representation and Decomposition in Safe Reinforcement Learning (NeurIPS 2024 · similarity 0.93)
- Safety Representations for Safer Policy Learning (ICLR 2025 · similarity 0.89)
- Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration (AAAI 2024 · similarity 0.89)
- Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization (NeurIPS 2025 · similarity 0.88)
- Handling Long and Richly Constrained Tasks through Constrained Hierarchical Reinforcement Learning (AAAI 2024 · similarity 0.88)
- Latent Safety-Constrained Policy Approach for Safe Offline Reinforcement Learning (ICLR 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
