mlsanityData to 14 Oct 2026

Interpretable and logic-guided RL

195 papers0 this weekLow confidence in this grouping: the papers may belong to more than one topic, or the group may changeLow confidence in this grouping: the papers may belong to more than one topic, or the group may change · About

Keywords: reinforcement, reinforcement learning, reward, symbolic, agent

Key facts. Interpretable and logic-guided RL: 195 papers on arXiv since 2024 (data through 14 Oct 2026). 13 of its 195 mapped papers (7%) cite “DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RL” (2024, arXiv 2410.04631). 3 new papers 30 Sep–14 Oct.

What’s new

30 Sep – 14 Oct 2026

3 new since 30 Sep

  1. Visualizing the World Model in an Agent's Value Function also fits: Deep RL algorithms and representations
  2. Towards the Automatic Synthesis of Interpretable Chess Tactics
  3. Minimal Witness Reinforcement Learning also fits: Deep RL algorithms and representations

Foundations

Most cited by papers here

  1. DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RL2024cited by 13 here
  2. Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learning2024cited by 12 here
  3. Interpretable Concept Bottlenecks to Align Reinforcement Learning Agents2024cited by 11 here
  4. LTL-Constrained Policy Optimization with Cycle Experience Replay2024cited by 8 here
  5. End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanations2024cited by 7 here

Lineage

The papers this direction builds on, oldest first.

Each paper on the line cites the one before.

  1. 2024Excluding the Irrelevant: Focusing Reinforcement Learning through Continuous Action Maskingcited by 2 here
  2. 2024Directed Exploration in Reinforcement Learning from Linear Temporal Logiccited by 6 here
  3. 2024DeepLTL: Learning to Efficiently Satisfy Complex LTL Specifications for Multi-Task RLcited by 13 here
  4. 2025One Subgoal at a Time: Zero-Shot Generalization to Arbitrary Linear Temporal Logic Requirements in Multi-Task Reinforcement Learningcited by 6 here
20 more it builds on, not in this chain
  1. 2024Interpretable Concept Bottlenecks to Align Reinforcement Learning Agentscited by 11 here
  2. 2024Symbolic Equation Solving via Reinforcement Learningcited by 2 here
  3. 2024On the Limitations of Markovian Rewards to Express Multi-Objective, Risk-Sensitive, and Modal Taskscited by 2 here
  4. 2024Logical Specifications-guided Dynamic Task Sampling for Reinforcement Learning Agentscited by 3 here
  5. 2024ACTER: Diverse and Actionable Counterfactual Sequences for Explaining and Diagnosing RL Policiescited by 2 here
  6. 2024End-to-End Neuro-Symbolic Reinforcement Learning with Textual Explanationscited by 7 here
  7. 2024LTL-Constrained Policy Optimization with Cycle Experience Replaycited by 8 here
  8. 2024Reinforcement learningcited by 7 here
  9. 2024Interpretable and Editable Programmatic Tree Policies for Reinforcement Learningcited by 6 here
  10. 2024Reward Machines for Deep RL in Noisy and Uncertain Environmentscited by 6 here
  11. 2024HackAtari: Atari Learning Environments for Robust and Continual Reinforcement Learningcited by 3 here
  12. 2024Gymnasium: A Standard Interface for Reinforcement Learning Environmentscited by 21 here
  13. 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 4 here
  14. 2024Reinforcement Learning with LTL and ω-Regular Objectives via Optimality-Preserving Translation to Average Rewardscited by 7 here
  15. 2024Compositional Automata Embeddings for Goal-Conditioned Reinforcement Learningcited by 12 here
  16. 2024Exploiting Hybrid Policy in Reinforcement Learning for Interpretable Temporal Logic Manipulationcited by 3 here
  17. 2024FORM: Learning Expressive and Transferable First-Order Logic Reward Machinescited by 4 here
  18. 2025Online inductive learning from answer sets for efficient reinforcement learning explorationcited by 3 here
  19. 2025A Survey on Explainable Deep Reinforcement Learningcited by 3 here
  20. 2025Pushdown Reward Machines for Reinforcement Learningcited by 3 here

The 24 most cited of 130.

Browse

30 of 195, ranked 6–35

  1. 6Reinforcement Learning with LTL and ω-Regular Objectives via Optimality-Preserving Translation to Average Rewards20247
  2. 7One Subgoal at a Time: Zero-Shot Generalization to Arbitrary Linear Temporal Logic Requirements in Multi-Task Reinforcement Learning20256
  3. 8Interpretable and Editable Programmatic Tree Policies for Reinforcement Learning20246
  4. 9Reward Machines for Deep RL in Noisy and Uncertain Environments20246
  5. 10Directed Exploration in Reinforcement Learning from Linear Temporal Logic20246
  6. 11FORM: Learning Expressive and Transferable First-Order Logic Reward Machines20244
  7. 12Online inductive learning from answer sets for efficient reinforcement learning exploration20253
  8. 13A Survey on Explainable Deep Reinforcement Learning20253
  9. 14Pushdown Reward Machines for Reinforcement Learning20253
  10. 15Logical Specifications-guided Dynamic Task Sampling for Reinforcement Learning Agents20243
All 30
  1. 16Exploiting Hybrid Policy in Reinforcement Learning for Interpretable Temporal Logic Manipulation20243
  2. 17PlatoLTL: Scaling LTL-Guided Multi-Task RL20262
  3. 18Semantically Labelled Automata for Multi-Task Reinforcement Learning with LTL Instructions20262
  4. 19Objective Metrics for Human-Subjects Evaluation in Explainable Reinforcement Learning20252
  5. 20Provably Correct Automata Embeddings for Optimal Automata-Conditioned Reinforcement Learning20252
  6. 21Evaluating Interpretable Reinforcement Learning by Distilling Policies into Programs20252
  7. 22A Theoretical Framework for Explaining Reinforcement Learning with Shapley Values20252
  8. 23Learning from Less: Guiding Deep Reinforcement Learning with Differentiable Symbolic Planning20252
  9. 24Reinforcement Learning with Stochastic Reward Machines20252
  10. 25Symbolic Equation Solving via Reinforcement Learning20242
  11. 26ACTER: Diverse and Actionable Counterfactual Sequences for Explaining and Diagnosing RL Policies20242
  12. 27EXPIL: Explanatory Predicate Invention for Learning in Games20242
  13. 28Neural Reward Machines20242
  14. 29Mitigating Information Loss in Tree-Based Reinforcement Learning via Direct Optimization20242
  15. 30Learning Robust Reward Machines from Noisy Labels20242
  16. 31Sample-Efficient Reinforcement Learning with Temporal Logic Objectives: Leveraging the Task Specification to Guide Exploration20242
  17. 32Interpretable end-to-end Neurosymbolic Reinforcement Learning agents20242
  18. 33Sample-Efficient Neurosymbolic Deep Reinforcement Learning20261
  19. 34Grounding LTL Tasks in Sub-Symbolic RL Environments for Zero-Shot Generalization20261
  20. 35Zero-Shot Instruction Following in RL via Structured LTL Representations20261