Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers

• URL: http://arxiv.org/abs/2606.21562v1
• Date: Fri, 19 Jun 2026 15:58:36 GMT
• Title: Compressing Observation History into Agent Memory: Distilling Transformers into Recurrent Transformers
• Abstract summary: We focus on long-horizon streaming vision and robotics applications like map-free pose estimation.<n>We argue that this gap does not stem from architectural limitations, but from differences in how these models learn to compress past information.<n>We show that this approach allows to train a recurrent latent robotic memory with linear-time complexity while substantially narrowing the performance gap to full-history transformers.
• Score: 25.583827940095322
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Transformers are AI's workhorse with strong performance in modeling sequential data, but their computational cost becomes prohibitive when processing long sequences. We target long-horizon streaming vision and robotics applications like map-free pose estimation, where it is particularly impractical to store and maintain a history of observations. Recurrent Transformers address this limitation by maintaining fixed-size memory but their performance lags behind that of transformers operating over the full observation history. We argue that this gap does not stem from architectural limitations, but from differences in how these models learn to compress past information. Without access to an observation history, recurrent models must explicitly decide what to retain in memory at each step, a significantly harder learning problem. In this work, we propose a distillation approach that transfers the compression strategy of a classical full-history transformer to a recurrent variant. We enable this by designing a teacher model that explicitly compresses its observation history into a fixed-size bottleneck representation. By directly supervising the student's memory with this bottleneck representation, we align the two compression mechanisms. We show that this approach allows to train a recurrent latent robotic memory with linear-time complexity while substantially narrowing the performance gap to full-history transformers.

Related papers

• MemoryWAM: Efficient World Action Modeling with Persistent Memory [80.90899269062128]
  We introduce MemoryWAM, a world action model with efficient persistent memory.<n>A tailored attention mechanism enables retrieval of both detailed short-term context and compressed long-term context.<n>MemoryWAM outperforms strong vision-language-action (VLA) and WAM baselines in both simulation and the real world.
  arXiv  Detail & Related papers  (2026-06-18T17:59:51Z)
• Memory Caching: RNNs with Growing Memory [56.25483647131372]
  We introduce Memory Caching (MC), a technique that enhances recurrent models by caching checkpoints of memory states (a.k.a. hidden states)<n>We propose four variants of MC, including gated aggregation and sparse selective mechanisms, and discuss their implications on both linear and deep memory modules.<n>The results indicate that while Transformers achieve the best accuracy, our MC variants show competitive performance, close the gap with Transformers, and performs better than state-of-the-art recurrent models.
  arXiv  Detail & Related papers  (2026-02-27T18:53:41Z)
• Kinaema: a recurrent sequence model for memory and pose in motion [25.677034515160567]
  We introduce a new model, Kinaema, and agent, capable of integrating a stream of visual observations while moving in a potentially large scene.<n>Our model does not explicitly store an observation history, therefore does not have hard constraints on context length.<n>We show that our large-capacity recurrent model maintains a useful representation of the scene, navigates to goals observed before the actual episode start, and is computationally efficient.
  arXiv  Detail & Related papers  (2025-10-23T06:34:53Z)
• Memo: Training Memory-Efficient Embodied Agents with Reinforcement Learning [53.72709564555407]
  Memo is a transformer-based architecture and training recipe for reinforcement learning.<n>It incorporates the creation and retrieval of memory by interleaving periodic summarization tokens with the inputs of a model during training.<n>We demonstrate Memo's effectiveness on a gridworld meta-RL benchmark and a multi-object navigation task in photo-realistic indoor settings.
  arXiv  Detail & Related papers  (2025-10-22T16:24:47Z)
• Recurrent Action Transformer with Memory [39.58317527488534]
  This paper proposes a novel model architecture that incorporates a recurrent memory mechanism designed to regulate information retention. We conduct experiments on memory-intensive environments (ViZDoom-Two-Colors, T-Maze, Memory Maze, Minigrid-Memory), classic Atari games, and MuJoCo control environments. The results show that using memory can significantly improve performance in memory-intensive environments, while maintaining or improving results in classic environments.
  arXiv  Detail & Related papers  (2023-06-15T19:29:08Z)
• AttMEMO : Accelerating Transformers with Memoization on Big Memory Systems [10.585040856070941]
  We introduce a novel embedding technique to find semantically similar inputs to identify computation similarity. We enable 22% inference-latency reduction on average (up to 68%) with negligible loss in inference accuracy.
  arXiv  Detail & Related papers  (2023-01-23T04:24:26Z)
• Transformers learn in-context by gradient descent [58.24152335931036]
  Training Transformers on auto-regressive objectives is closely related to gradient-based meta-learning formulations. We show how trained Transformers become mesa-optimizers i.e. learn models by gradient descent in their forward pass.
  arXiv  Detail & Related papers  (2022-12-15T09:21:21Z)
• Token Turing Machines [53.22971546637947]
  Token Turing Machines (TTM) is a sequential, autoregressive Transformer model with memory for real-world sequential visual understanding. Our model is inspired by the seminal Neural Turing Machine, and has an external memory consisting of a set of tokens which summarise the previous history.
  arXiv  Detail & Related papers  (2022-11-16T18:59:18Z)
• Recurrent Memory Transformer [0.3529736140137003]
  We study a memory-augmented segment-level recurrent Transformer (Recurrent Memory Transformer) We implement a memory mechanism with no changes to Transformer model by adding special memory tokens to the input or output sequence. Our model performs on par with the Transformer-XL on language modeling for smaller memory sizes and outperforms it for tasks that require longer sequence processing.
  arXiv  Detail & Related papers  (2022-07-14T13:00:22Z)
• Memformer: A Memory-Augmented Transformer for Sequence Modeling [55.780849185884996]
  We present Memformer, an efficient neural network for sequence modeling. Our model achieves linear time complexity and constant memory space complexity when processing long sequences.
  arXiv  Detail & Related papers  (2020-10-14T09:03:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.