aiqbench analysis
PHASER addresses catastrophic forgetting in Vision-Language-Action models by replacing naive uniform sampling with a phase-centric memory allocation strategy that prioritizes critical sub-skills at risk of decay. The framework achieves autonomous lifelong adaptation through an unsupervised pipeline that detects temporal boundaries via action-signal changes and verifies them semantically. Evaluated on standard robotic suites, this architecture-agnostic approach significantly boosts average success rates while maintaining balanced retention across diverse tasks.
PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Models
arXiv:2606.03598v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have achieved remarkable success in language-conditioned robotic manipulation. However, deploying these models in open-ended environments requires continuously acquiring novel skills, a process that inevitably triggers severe catastrophic forgetting of previously learned behaviors. While experience replay (ER) serves as a standard mitigating strategy, naive uniform sampling fundamentally misaligns with the temporal characteristics of manipulation trajectories. It systematically under-samples brief but causally critical sub-skills, leading to phase starvation, and completely overlooks the varying degrees of forgetting across historical tasks. To overcome these limitations, we introduce PHASER, an architecture-agnostic continual learning framework. PHASER employs a phase-centric capacity allocation to guarantee equal memory support for all sub-skills, coupled with a multi-modal interference routing strategy that dynamically prioritizes historical phases at high risk of forgetting. Furthermore, to enable fully autonomous lifelong adaptation, we integrate Auto-PC, a lightweight pipeline combining unsupervised action-signal change-point detection with VLM-based semantic verification to extract temporal boundaries without intensive manual supervision. Evaluated across three VLA backbones on LIBERO continual learning suites, PHASER yields substantial empirical improvements, increasing Average Success Rate (ASR) by up to 31% over matched-budget ER and achieving an 87.8% final ASR on the LIBERO-Goal CL setting.
Abstract:Vision-Language-Action (VLA) models have achieved remarkable success in language-conditioned robotic manipulation. However, deploying these models in open-ended environments requires continuously acquiring novel skills, a process that inevitably triggers severe catastrophic forgetting of previously learned behaviors. While experience replay (ER) serves as a standard mitigating strategy, naive uniform sampling fundamentally misaligns with the temporal characteristics of manipulation trajectories. It systematically under-samples brief but causally critical sub-skills, leading to phase starvation, and completely overlooks the varying degrees of forgetting across historical tasks. To overcome these limitations, we introduce PHASER, an architecture-agnostic continual learning framework. PHASER employs a phase-centric capacity allocation to guarantee equal memory support for all sub-skills, coupled with a multi-modal interference routing strategy that dynamically prioritizes historical phases at high risk of forgetting. Furthermore, to enable fully autonomous lifelong adaptation, we integrate Auto-PC, a lightweight pipeline combining unsupervised action-signal change-point detection with VLM-based semantic verification to extract temporal boundaries without intensive manual supervision. Evaluated across three VLA backbones on LIBERO continual learning suites, PHASER yields substantial empirical improvements, increasing Average Success Rate (ASR) by up to 31% over matched-budget ER and achieving an 87.8% final ASR on the LIBERO-Goal CL setting.
Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2606.03598 [cs.RO]
(or arXiv:2606.03598v2 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.03598 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 291 words
Related coverage
PHASER: Phase-Aware and Semantic Experience Replay for Vision-Language-Action Models
arXiv · Jun 4, 2026
Towards Human-like Physical Intelligence: Lifelong Vision-Language-Action Learning for Robotic Manipulation
arXiv · Jul 22, 2026
Building a Foundation Stack for General-Purpose Robots
IEEE Spectrum · Jul 13, 2026
What's Continual Learning, and Why Might We Expect To See It In Advanced LLM Agents?
LessWrong · Jun 12, 2026