Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning

• URL: http://arxiv.org/abs/2603.03818v1
• Date: Wed, 04 Mar 2026 08:03:13 GMT
• Title: Pretrained Vision-Language-Action Models are Surprisingly Resistant to Forgetting in Continual Learning
• Abstract summary: We study the behavior of large-scale pretrained Vision-Language-Action (VLA) models trained from scratch.<n>We find that pretrained VLAs are remarkably resistant to forgetting compared with smaller policy models trained from scratch.<n>Our analysis reveals that pretraining plays a critical role in downstream continual learning performance.
• Score: 30.941068379610794
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Continual learning is a long-standing challenge in robot policy learning, where a policy must acquire new skills over time without catastrophically forgetting previously learned ones. While prior work has extensively studied continual learning in relatively small behavior cloning (BC) policy models trained from scratch, its behavior in modern large-scale pretrained Vision-Language-Action (VLA) models remains underexplored. In this work, we found that pretrained VLAs are remarkably resistant to forgetting compared with smaller policy models trained from scratch. Simple Experience Replay (ER) works surprisingly well on VLAs, sometimes achieving zero forgetting even with a small replay data size. Our analysis reveals that pretraining plays a critical role in downstream continual learning performance: large pretrained models mitigate forgetting with a small replay buffer size while maintaining strong forward learning capabilities. Furthermore, we found that VLAs can retain relevant knowledge from prior tasks despite performance degradation during learning new tasks. This knowledge retention enables rapid recovery of seemingly forgotten skills through finetuning. Together, these insights imply that large-scale pretraining fundamentally changes the dynamics of continual learning, enabling models to continually acquire new skills over time with simple replay. Code and more information can be found at https://ut-austin-rpl.github.io/continual-vla

Related papers

• When Does Continual Learning Require Learning [68.55508958298942]
  We show that continual learning is not a single capability.<n>Different patterns of environmental change require fundamentally different update behaviors.<n>We hope that understanding where each method succeeds and fails will guide the design of stronger continual learning systems.
  arXiv  Detail & Related papers  (2026-07-08T18:27:40Z)
• Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay [44.18996691373878]
  We show that self-generated samples serve as effective replay data, nearly eliminating forgetting.<n>We find that forgetting persists when the model has little remaining capacity.
  arXiv  Detail & Related papers  (2026-05-25T17:54:34Z)
• STAR: Stability-Inducing Weight Perturbation for Continual Learning [4.623295991242981]
  A key challenge in continual learning is that as the model is updated to learn new tasks, it becomes susceptible to catastrophic forgetting.<n>A popular approach to mitigate forgetting during continual learning is to maintain a small buffer of previously-seen samples and to replay them during training.<n>We propose a novel loss function, STAR, that exploits the worst-case parameter perturbation that reduces the KL-divergence of model predictions.
  arXiv  Detail & Related papers  (2025-03-03T14:32:03Z)
• Merge to Learn: Efficiently Adding Skills to Language Models with Model Merging [102.16497861225358]
  Adapting general-purpose language models to new skills is currently an expensive process. We investigate the effectiveness of adding new skills to preexisting models by training on the new skills in isolation and later merging with the general model.
  arXiv  Detail & Related papers  (2024-10-16T18:23:50Z)
• Beyond Prompt Learning: Continual Adapter for Efficient Rehearsal-Free Continual Learning [22.13331870720021]
  We propose a beyond prompt learning approach to the RFCL task, called Continual Adapter (C-ADA) C-ADA flexibly extends specific weights in CAL to learn new knowledge for each task and freezes old weights to preserve prior knowledge. Our approach achieves significantly improved performance and training speed, outperforming the current state-of-the-art (SOTA) method.
  arXiv  Detail & Related papers  (2024-07-14T17:40:40Z)
• Replay-enhanced Continual Reinforcement Learning [37.34722105058351]
  We introduce RECALL, a replay-enhanced method that greatly improves the plasticity of existing replay-based methods on new tasks. Experiments on the Continual World benchmark show that RECALL performs significantly better than purely perfect memory replay.
  arXiv  Detail & Related papers  (2023-11-20T06:21:52Z)
• Towards Robust Continual Learning with Bayesian Adaptive Moment Regularization [51.34904967046097]
  Continual learning seeks to overcome the challenge of catastrophic forgetting, where a model forgets previously learnt information. We introduce a novel prior-based method that better constrains parameter growth, reducing catastrophic forgetting. Results show that BAdam achieves state-of-the-art performance for prior-based methods on challenging single-headed class-incremental experiments.
  arXiv  Detail & Related papers  (2023-09-15T17:10:51Z)
• Detachedly Learn a Classifier for Class-Incremental Learning [11.865788374587734]
  We present an analysis that the failure of vanilla experience replay (ER) comes from unnecessary re-learning of previous tasks and incompetence to distinguish current task from the previous ones. We propose a novel replay strategy task-aware experience replay. Experimental results show our method outperforms current state-of-the-art methods.
  arXiv  Detail & Related papers  (2023-02-23T01:35:44Z)
• PIVOT: Prompting for Video Continual Learning [50.80141083993668]
  We introduce PIVOT, a novel method that leverages extensive knowledge in pre-trained models from the image domain. Our experiments show that PIVOT improves state-of-the-art methods by a significant 27% on the 20-task ActivityNet setup.
  arXiv  Detail & Related papers  (2022-12-09T13:22:27Z)
• PEBBLE: Feedback-Efficient Interactive Reinforcement Learning via Relabeling Experience and Unsupervised Pre-training [94.87393610927812]
  We present an off-policy, interactive reinforcement learning algorithm that capitalizes on the strengths of both feedback and off-policy learning. We demonstrate that our approach is capable of learning tasks of higher complexity than previously considered by human-in-the-loop methods.
  arXiv  Detail & Related papers  (2021-06-09T14:10:50Z)
• Lifelong Learning of Few-shot Learners across NLP Tasks [45.273018249235705]
  We study the challenge of lifelong learning to few-shot learn over a sequence of diverse NLP tasks. We propose a continual meta-learning approach which learns to generate adapter weights from a few examples. We demonstrate our approach preserves model performance over training tasks and leads to positive knowledge transfer when the future tasks are learned.
  arXiv  Detail & Related papers  (2021-04-18T10:41:56Z)
• Parrot: Data-Driven Behavioral Priors for Reinforcement Learning [79.32403825036792]
  We propose a method for pre-training behavioral priors that can capture complex input-output relationships observed in successful trials. We show how this learned prior can be used for rapidly learning new tasks without impeding the RL agent's ability to try out novel behaviors.
  arXiv  Detail & Related papers  (2020-11-19T18:47:40Z)
• Bilevel Continual Learning [76.50127663309604]
  We present a novel framework of continual learning named "Bilevel Continual Learning" (BCL) Our experiments on continual learning benchmarks demonstrate the efficacy of the proposed BCL compared to many state-of-the-art methods.
  arXiv  Detail & Related papers  (2020-07-30T16:00:23Z)

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.