# Automatic Recall Machines: Internal Replay, Continual Learning and the Brain

arXiv: https://arxiv.org/abs/2006.12323 · HTML: https://mlsanity.com/paper/2006.12323/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2020.

## Built on by (most cited first, 7 of 7)

- [New Insights on Reducing Abrupt Representation Change in Online Continual Learning](https://mlsanity.com/paper/2104.05025/) (2021)
- [Always Be Dreaming: A New Approach for Data-Free Class-Incremental Learning](https://mlsanity.com/paper/2106.09701/) (2021)
- [Prototype-Sample Relation Distillation: Towards Replay-Free Continual Learning](https://mlsanity.com/paper/2303.14771/) (2023)
- [On the importance of cross-task features for class-incremental learning](https://mlsanity.com/paper/2106.11930/) (2021)
- [CRAM: Large-scale Video Continual Learning with Bootstrapped Compression](https://mlsanity.com/paper/2508.05001/) (2025)
- [New Insights on Reducing Abrupt Representation Change in Online Continual Learning](https://mlsanity.com/paper/2203.03798/) (2022)
- [PVBF: A Framework for Mitigating Parameter Variation Imbalance in Online Continual Learning](https://mlsanity.com/paper/2502.17794/) (2025)

## Abstract (arXiv)

Replay in neural networks involves training on sequential data with memorized samples, which counteracts forgetting of previous behavior caused by non-stationarity. We present a method where these auxiliary samples are generated on the fly, given only the model that is being trained for the assessed objective, without extraneous buffers or generator networks. Instead the implicit memory of learned samples within the assessed model itself is exploited. Furthermore, whereas existing work focuses on reinforcing the full seen data distribution, we show that optimizing for not forgetting calls for the generation of samples that are specialized to each real training batch, which is more efficient and scalable. We consider high-level parallels with the brain, notably the use of a single model for inference and recall, the dependency of recalled samples on the current environment batch, top-down modulation of activations and learning, abstract recall, and the dependency between the degree to which a task is learned and the degree to which it is recalled. These characteristics emerge naturally from the method without being controlled for.