mlsanityData to 7 Oct 2026

SpecRLBench: A Benchmark for Generalization in Specification-Guided Reinforcement Learning

arXiv 2604.24729 ↗First posted 27 Apr 2026cs.LGcited by 1

Not on the map. Why?

Abstract (arXiv)

Specification-guided reinforcement learning (RL) provides a principled framework for encoding complex, temporally extended tasks using formal specifications such as linear temporal logic (LTL). While recent methods have shown promising results, their ability to generalize across unseen specifications and diverse environments remains insufficiently understood. In this work, we introduce SpecRLBench, a benchmark designed to evaluate the generalization capabilities of LTL-based specification-guided RL methods. The benchmark spans multiple difficulty levels across navigation and manipulation domains, incorporating both static and dynamic environments, diverse robot dynamics, and varied observation modalities. Through extensive empirical evaluation, we characterize the strengths and limitations of existing approaches and reveal the challenges that emerge as specification and environment complexity increase. SpecRLBench provides a structured platform for systematic comparison and supports the development of more generalizable specification-guided RL methods. Code is available at https://github.com/BU-DEPEND-Lab/SpecRLBench.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 55 of its 72 references that are in the map. First posted on arXiv 27 Apr 2026.

Builds on

55 of 72 references.

References · 55

  1. 2020D4RL: Datasets for Deep Data-Driven Reinforcement Learningcited by 1,406
  2. 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
  3. 2019Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learningcited by 1,168
  4. 2025Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Successcited by 844
  5. 2021CALVIN: A Benchmark for Language-Conditioned Policy Learning for Long-Horizon Robot Manipulation Taskscited by 564
  6. 2023Minigrid & Miniworld: Modular & Customizable Reinforcement Learning Environments for Goal-Oriented Taskscited by 320
  7. 2018BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learningcited by 238
  8. 2024OGBench: Benchmarking Offline Goal-Conditioned RLcited by 216
  9. 2020Reward Machines: Exploiting Reward Function Structure in Reinforcement Learningcited by 183
  10. 2024Precise and Dexterous Robotic Manipulation via Human-in-the-Loop Reinforcement Learningcited by 153
  11. 2022Goal-Conditioned Reinforcement Learning: Problems and Solutionscited by 145
  12. 2016Reinforcement Learning With Temporal Logic Rewardscited by 114
38 more
  1. 2018Omega-Regular Objectives in Model-Free Reinforcement Learningcited by 74
  2. 2019Control Synthesis from Linear Temporal Logic Specifications using Model-Free Reinforcement Learningcited by 74
  3. 2021LTL2Action: Generalizing LTL Instructions for Multi-Task RLcited by 72
  4. 2018Foundations for Restraining Bolts: Reinforcement Learning with LTLf/LDLf restraining specificationscited by 69
  5. 2020A Composable Specification Language for Reinforcement Learning Taskscited by 68
  6. 2016Q-Learning for Robust Satisfaction of Signal Temporal Logic Specificationscited by 65
  7. 2021Compositional Reinforcement Learning from Logical Specificationscited by 65
  8. 2021panda-gym: Open-source goal-conditioned environments for robotic learningcited by 56
  9. 2025RLinf: Flexible and Efficient Large-scale Reinforcement Learning via Macro-to-Micro Flow Transformationcited by 51
  10. 2024VLM-RL: A Unified Vision Language Models and Reinforcement Learning Framework for Safe Autonomous Drivingcited by 49
  11. 2021CORA: Benchmarks, Baselines, and Metrics as a Platform for Continual Reinforcement Learning Agentscited by 36
  12. 2025CarPlanner: Consistent Auto-regressive Trajectory Planning for Large-scale Reinforcement Learning in Autonomous Drivingcited by 31
  13. 2021Multi-Agent Reinforcement Learning with Temporal Logic Specificationscited by 27
  14. 2021The Logical Options Frameworkcited by 24
  15. 2020Safe Reinforcement Learning with Natural Language Constraintscited by 22
  16. 2022CompoSuite: A Compositional Reinforcement Learning Benchmarkcited by 21
  17. 2024DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RLcited by 20
  18. 2023Eventual Discounting Temporal Logic Counterfactual Experience Replaycited by 18
  19. 2024Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learningcited by 18
  20. 2023Sample Efficient Model-free Reinforcement Learning from LTL Specifications with Optimality Guaranteescited by 17
  21. 2024Reinforcement Learning with LTL and ω-Regular Objectives via Optimality-Preserving Translation to Average Rewardscited by 13
  22. 2022Overcoming Exploration: Deep Reinforcement Learning for Continuous Control in Cluttered Environments from Temporal Logic Specificationscited by 11
  23. 2024Temporal Logic Specification-Conditioned Decision Transformer for Offline Safe Reinforcement Learningcited by 10
  24. 2024LTL-Constrained Policy Optimization with Cycle Experience Replaycited by 10
  25. 2022A Hierarchical Bayesian Approach to Inverse Reinforcement Learning with Symbolic Reward Machinescited by 9
  26. 2025TeLoGraF: Temporal Logic Planning via Graph-encoded Flow Matchingcited by 9
  27. 2025One Subgoal at a Time: Zero-Shot Generalization to Arbitrary Linear Temporal Logic Requirements in Multi-Task Reinforcement Learningcited by 8
  28. 2022Skill Machines: Temporal Logic Skill Composition in Reinforcement Learningcited by 7
  29. 2024Directed Exploration in Reinforcement Learning from Linear Temporal Logiccited by 7
  30. 2023Multi-Agent Reinforcement Learning Guided by Signal Temporal Logic Specificationscited by 6
  31. 2024Learning Whole-Body Loco-Manipulation for Omni-Directional Task Space Pose Tracking with a Wheeled-Quadrupedal-Manipulatorcited by 5
  32. 2025TGPO: Temporal Grounded Policy Optimization for Signal Temporal Logic Taskscited by 5
  33. 2022DistSPECTRL: Distributing Specifications in Multi-Agent Reinforcement Learning Systemscited by 3
  34. 2024Logical Specifications-guided Dynamic Task Sampling for Reinforcement Learning Agentscited by 3
  35. 2025GoalLadder: Incremental Goal Discovery with Vision-Language Modelscited by 3
  36. 2023Learning Reward Machines in Cooperative Multi-Agent Taskscited by 2
  37. 2024GLIDE-RL: Grounded Language Instruction through DEmonstration in RLcited by 2
  38. 2025ImagineBench: Evaluating Reinforcement Learning with Large Language Model Rolloutscited by 2

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Jaxolotl: A Unified High-Performance Benchmark Suite for LTL-Based Multi-Task RLcited by 0