mlsanityData to 13 Oct 2026

LANTERN: LLM-Augmented Neurosymbolic Transfer with Experience-Gated Reasoning Networks

arXiv 2605.05478 ↗First posted 6 May 2026cs.AIcited by 0

On the map: LLM reasoning → Reinforcement learning for reasoning

Abstract (arXiv)

Transfer learning in reinforcement learning (RL) seeks to accelerate learning in new tasks by leveraging knowledge from related sources. Existing neurosymbolic transfer methods, however, typically rely on manually specified task automata, assume a single source task, and use fixed knowledge-integration mechanisms that cannot adapt to varying source relevance. We propose LANTERN, a unified framework for multi-source neurosymbolic transfer that addresses these limitations through three components: (i) deterministic finite automata generated from natural language task descriptions using large language models, (ii) semantic embedding-based aggregation of multiple source policies weighted by cross-task similarity, and (iii) adaptive teacher-student gating based on temporal-difference error and semantic uncertainty. Across domains spanning resource management, navigation, and control, LANTERN achieves 40-60% improvements in sample efficiency over existing baselines while remaining robust to poorly aligned sources. These results demonstrate that multi-source, adaptively weighted neurosymbolic transfer can improve scalability and robustness in symbolic RL settings.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 23 of its 40 references that are in the map. First posted on arXiv 6 May 2026.

Builds on

23 of 40 references.

References · 23

  1. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,870
  2. 2015End-to-End Training of Deep Visuomotor Policiescited by 850
  3. 2020Deep Reinforcement Learning for Autonomous Driving: A Surveycited by 699
  4. 2015Policy Distillationcited by 423
  5. 2019Efficient Off-Policy Meta-Reinforcement Learning via Probabilistic Context Variablescited by 373
  6. 2016Successor Features for Transfer in Reinforcement Learningcited by 332
  7. 2019Challenges of Real-World Reinforcement Learningcited by 285
  8. 2020Transfer Learning in Deep Reinforcement Learning: A Surveycited by 259
  9. 2017Distral: Robust Multitask Reinforcement Learningcited by 231
  10. 2016Modular Multitask Reinforcement Learning with Policy Sketchescited by 200
  11. 2018Programmatically Interpretable Reinforcement Learningcited by 192
  12. 2020Reward Machines: Exploiting Reward Function Structure in Reinforcement Learningcited by 183
11 more
  1. 2015Actor-Mimic: Deep Multitask and Transfer Reinforcement Learningcited by 177
  2. 2019Language as an Abstraction for Hierarchical Deep Reinforcement Learningcited by 138
  3. 2017Zero-Shot Task Generalization with Multi-Task Deep Reinforcement Learningcited by 97
  4. 2019Distilling Policy Distillationcited by 83
  5. 2018Omega-Regular Objectives in Model-Free Reinforcement Learningcited by 74
  6. 2018Foundations for Restraining Bolts: Reinforcement Learning with LTLf/LDLf restraining specificationscited by 69
  7. 2017Environment-Independent Task Specifications via GLTLcited by 67
  8. 2020Cautious Reinforcement Learning with Logical Constraintscited by 44
  9. 2025ARM-FM: Automated Reward Machines via Foundation Models for Compositional Reinforcement Learningcited by 5
  10. 2023Automaton Distillation: Neuro-Symbolic Transfer Learning for Deep Reinforcement Learningcited by 4
  11. 2026CADENT: Gated Hybrid Distillation for Sample-Efficient Transfer in Reinforcement Learningcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Knowledge is Not Enough: Injecting RL Skills for Continual Adaptation
  2. 2025Knowledge Transfer in Model-Based Reinforcement Learning Agents for Efficient Multi-Task Learning
  3. 2023Foundations for Transfer in Reinforcement Learning: A Taxonomy of Knowledge Modalities
  4. 2026ASPECT:Analogical Semantic Policy Execution via Language Conditioned Transfer
  5. 2026Neuro-Symbolic Injection of LTLf Constraints in Autoregressive Reinforcement Learning Policies
  6. 2025WST: Weak-to-Strong Knowledge Transfer via Reinforcement Learning
  7. 2026From Passive Reuse to Active Reasoning: Grounding Large Language Models for Neuro-Symbolic Experience Replay
  8. 2026Experience Augmented Policy Optimization for LLM Reasoning