mlsanityData to 14 Oct 2026

Latent learning: episodic memory complements parametric learning by enabling flexible reuse of experiences

arXiv 2509.16189 ↗First posted 19 Sep 2025cs.LG cs.CLcited by 15

Not on the map. Why?

Abstract (arXiv)

When do machine learning systems fail to generalize, and what mechanisms could improve their generalization? Here, we draw inspiration from cognitive science to argue that one weakness of parametric machine learning systems is their failure to exhibit latent learning -- learning information that is not relevant to the task at hand, but that might be useful in a future task. We show how this perspective links failures ranging from the reversal curse in language modeling to new findings on agent-based navigation. We then highlight how cognitive science points to episodic memory as a potential part of the solution to these issues. Correspondingly, we show that a system with an oracle retrieval mechanism can use learning experiences more flexibly to generalize better across many of these challenges. We also identify some of the essential components for effectively using retrieval, including the importance of within-example in-context learning for acquiring the ability to use information across retrieved examples. In summary, our results illustrate one possible contributor to the relative data inefficiency of current machine learning systems compared to natural intelligence, and help to understand how retrieval methods can complement parametric learning to improve generalization. We close by discussing some of the links between these findings and prior results in cognitive science and neuroscience, and the broader implications.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 42 of its 88 references that are in the map. First posted on arXiv 19 Sep 2025.

Builds on

42 of 88 references.

References · 42

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,182
  5. 2013Distributed Representations of Words and Phrases and their Compositionalitycited by 3,474
  6. 2021Improving language models by retrieving from trillions of tokenscited by 1,293
  7. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,178
  8. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  9. 2018IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architecturescited by 883
  10. 2023The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"cited by 407
  11. 2016Learning to reinforcement learncited by 377
  12. 2022Data Distributional Properties Drive Emergent In-Context Learning in Transformerscited by 309
30 more
  1. 2019Stabilizing Transformers for Reinforcement Learningcited by 270
  2. 2023Language Modeling Is Compressioncited by 225
  3. 2017Neural Episodic Controlcited by 151
  4. 2023Taken out of context: On measuring situational awareness in LLMscited by 143
  5. 2023Cross-Lingual Consistency of Factual Knowledge in Multilingual Language Modelscited by 118
  6. 2023Test-Time Training on Nearest Neighbors for Large Language Modelscited by 88
  7. 2024Language Models Learn Rare Phenomena from Less Rare Phenomena: The Case of the Missing AANNscited by 65
  8. 2021Relating transformers to models and neural representations of the hippocampal formationcited by 62
  9. 2020Grounded Language Learning Fast and Slowcited by 54
  10. 2022Retrieval-Augmented Reinforcement Learningcited by 53
  11. 2023Grokking of Hierarchical Structure in Vanilla Transformerscited by 53
  12. 2022Transformers generalize differently from information stored in context vs in weightscited by 49
  13. 2025Reasoning to Learn from Latent Thoughtscited by 45
  14. 2024Efficiently Learning at Test-Time: Active Fine-Tuning of LLMscited by 42
  15. 2021Towards mental time travel: a hierarchical memory for reinforcement learning agentscited by 40
  16. 2025On the generalization of language models from in-context learning and finetuning: a controlled studycited by 40
  17. 2024The broader spectrum of in-context learningcited by 35
  18. 2025ECLeKTic: a Novel Challenge Set for Evaluation of Cross-Lingual Knowledge Transfercited by 29
  19. 2018Been There, Done That: Meta-Learning with Episodic Recallcited by 26
  20. 2022Large-Scale Retrieval for Reinforcement Learningcited by 24
  21. 2025Key-value memory in the braincited by 24
  22. 2024Deductive Closure Training of Language Models for Coherence, Accuracy, and Updatabilitycited by 22
  23. 2023Passive learning of active causal strategies in agents and language modelscited by 21
  24. 2025Language Models' Factuality Depends on the Language of Inquirycited by 14
  25. 2022Zipfian environments for Reinforcement Learningcited by 13
  26. 2023Tell, don't show: Declarative facts influence how LLMs generalizecited by 9
  27. 2022Prospective Learning: Principled Extrapolation to the Futurecited by 6
  28. 2023Exploring the Reversal Curse and Other Deductive Logical Reasoning in BERT and GPT-Based Large Language Modelscited by 5
  29. 2025Preemptive Solving of Future Problems: Multitask Preplay in Humans and Machinescited by 2
  30. 2025A Taxonomy of Transcendencecited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 12.

15 papers.

Sort:

By area:

  1. 2026Learning Next Action Predictors from Human-Computer Interactioncited by 7
  2. 2026Contextual Agentic Memory is a Memo, Not True Memorycited by 6
  3. 2026Do Language Models Track Entities Across State Changes?cited by 6
  4. 2025Knowing the Answer Isn't Enough: Fixing Reasoning Path Failures in LVLMscited by 5
  5. 2025ReEXplore: Improving MLLMs for Embodied Exploration with Contextualized Retrospective Experience Replaycited by 4
  6. 2025Beyond Heuristics: A Decision-Theoretic Framework for Agent Memory Managementcited by 4
  7. 2026HIMM: Human-Inspired Long-Term Memory Modeling for Embodied Exploration and Question Answeringcited by 3
  8. 2026Organize then Retrieve: Hierarchical Memory Navigation for Efficient Agentscited by 3
  9. 2026Revolutionizing Long-Term Memory in AI: New Horizons with High-Capacity and High-Speed Storagecited by 1
  10. 2026Synthetic Mixed Training: Scaling Parametric Knowledge Acquisition Beyond RAGcited by 1
  11. 2026SRAG: RAG with Structured Data Improves Vector Retrievalcited by 0
  12. 2026Factorization Regret mediates compositional generalization in latent spacecited by 0
3 more
  1. 2026Improving Latent Generalization Using Test-time Computecited by 0
  2. 2026The Illusion of Latent Generalization: Bi-directionality and the Reversal Cursecited by 0
  3. 2026Does Episodic Memory Help Close the Lexical Frequency Gap in Sensitivity to Syntactic Contrasts? A Test Using Retrieval-Augmented Language Modelscited by 0