Maintain Plasticity in Long-timescale Continual Test-time Adaptation

• URL: http://arxiv.org/abs/2412.20034v1
• Date: Sat, 28 Dec 2024 05:47:40 GMT
• Title: Maintain Plasticity in Long-timescale Continual Test-time Adaptation
• Abstract summary: Continual test-time domain adaptation aims to adjust pre-trained source models to perform well over time across non-stationary target environments.<n>The plasticity refers to the model's capability to adjust predictions in response to non-stationary environments continually.<n>We propose a simple yet effective policy, Adaptive Shrink-Restore (ASR), towards preserving the model's plasticity.
• Score: 25.795708177210876
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Continual test-time domain adaptation (CTTA) aims to adjust pre-trained source models to perform well over time across non-stationary target environments. While previous methods have made considerable efforts to optimize the adaptation process, a crucial question remains: can the model adapt to continually-changing environments with preserved plasticity over a long time? The plasticity refers to the model's capability to adjust predictions in response to non-stationary environments continually. In this work, we explore plasticity, this essential but often overlooked aspect of continual adaptation to facilitate more sustained adaptation in the long run. First, we observe that most CTTA methods experience a steady and consistent decline in plasticity during the long-timescale continual adaptation phase. Moreover, we find that the loss of plasticity is strongly associated with the change in label flip. Based on this correlation, we propose a simple yet effective policy, Adaptive Shrink-Restore (ASR), towards preserving the model's plasticity. In particular, ASR does the weight re-initialization by the adaptive intervals. The adaptive interval is determined based on the change in label flipping. Our method is validated on extensive CTTA benchmarks, achieving excellent performance.

Related papers

• TTSD-FAR: Test-Time Self-Distillation with Fisher-Anchored Restoration for Missing-Modality Emotion Recognition in LVLMs [83.78708832901194]
  Large video-language models (LVLMs) have shown remarkable performance on multimodal tasks like multimodal emotion recognition (ER) in the wild.<n>However, real-world deployment remains challenging: modalities may be missing or noisy at test time.<n>We propose a solution in which a frozen teacher guides an adaptive low-rank student via self-distillation.<n>Experiments on MELD, DFEW, and BAH under 0%-50% missing modalities show that this unified adaptation-restoration design consistently outperforms entropy-based adaptation, RAG, and perplexity-based generation.
  arXiv  Detail & Related papers  (2026-08-18T23:40:28Z)
• PEFT-Arena: Understanding Parameter-Efficient Finetuning from a Stability-Plasticity Perspective [52.693471818837395]
  We argue that PEFT should be assessed through the stability-plasticity dilemma.<n>We introduce PEFT-Arena, a benchmark that measures downstream performance and general capability retention.<n>Inspired by this, we present case studies of a post-hoc improvement with path-wise rewinding.
  arXiv  Detail & Related papers  (2026-05-27T17:59:51Z)
• Balancing Plasticity and Stability with Fast and Slow Successor Features [29.71873686385453]
  We study how stability and adaptation affect performance under continuous environmental change.<n>We find that methods favoring stability, such as synaptic consolidation, outperform approaches focused on plasticity.
  arXiv  Detail & Related papers  (2026-05-25T22:05:32Z)
• Adaptive and Balanced Re-initialization for Long-timescale Continual Test-time Domain Adaptation [27.967679954595123]
  Continual test-time domain adaptation aims to adjust models so that they can perform well over time across non-stationary environments.<n>Can the model adapt to continually changing environments over a long time?<n>We propose Adaptive-and-Balanced Re-initialization (ABR) towards preserving the model's long-term performance.
  arXiv  Detail & Related papers  (2026-02-06T02:53:15Z)
• Not All Preferences Are Created Equal: Stability-Aware and Gradient-Efficient Alignment for Reasoning Models [52.48582333951919]
  We propose a dynamic framework designed to enhance alignment reliability by maximizing the Signal-to-Noise Ratio of policy updates.<n>SAGE (Stability-Aware Gradient Efficiency) integrates a coarse-grained curriculum mechanism that refreshes candidate pools based on model competence.<n> Experiments on multiple mathematical reasoning benchmarks demonstrate that SAGE significantly accelerates convergence and outperforms static baselines.
  arXiv  Detail & Related papers  (2026-02-01T12:56:10Z)
• SPRInG: Continual LLM Personalization via Selective Parametric Adaptation and Retrieval-Interpolated Generation [11.253466187411524]
  We introduce SPRInG, a novel semi-parametric framework designed for effective continual personalization.<n>During training, SPRInG employs drift-driven selective adaptation, which utilizes a likelihood-based scoring function to identify high-novelty interactions.<n>Experiments on the long-form personalized generation benchmark demonstrate that SPRInG outperforms existing baselines.
  arXiv  Detail & Related papers  (2026-01-15T01:32:27Z)
• GateRA: Token-Aware Modulation for Parameter-Efficient Fine-Tuning [51.79350934271497]
  GateRA is a unified framework that introduces token-aware modulation to dynamically adjust the strength of PEFT updates.<n>By incorporating adaptive gating into standard PEFT branches, GateRA enables selective, token-level adaptation.<n> Experiments on multiple commonsense reasoning benchmarks demonstrate that GateRA consistently outperforms or matches prior PEFT methods.
  arXiv  Detail & Related papers  (2025-11-15T17:55:47Z)
• Adapting to Fragmented and Evolving Data: A Fisher Information Perspective [0.0]
  FADE is a lightweight framework for robust learning under dynamic environments.<n>It employs a shift-aware regularization mechanism anchored in Fisher information geometry.<n>FADE operates online with fixed memory and no access to target labels.
  arXiv  Detail & Related papers  (2025-07-25T06:50:09Z)
• Orthogonal Projection Subspace to Aggregate Online Prior-knowledge for Continual Test-time Adaptation [67.80294336559574]
  Continual Test Time Adaptation (CTTA) is a task that requires a source pre-trained model to continually adapt to new scenarios.<n>We propose a novel pipeline, Orthogonal Projection Subspace to aggregate online Prior-knowledge, dubbed OoPk.
  arXiv  Detail & Related papers  (2025-06-23T18:17:39Z)
• CANet: ChronoAdaptive Network for Enhanced Long-Term Time Series Forecasting under Non-Stationarity [0.0]
  We introduce a novel architecture, ChoronoAdaptive Network (CANet), inspired by style-transfer techniques. The core of CANet is the Non-stationary Adaptive Normalization module, seamlessly integrating the Style Blending Gate and Adaptive Instance Normalization (AdaIN) experiments on real-world datasets validate CANet's superiority over state-of-the-art methods, achieving a 42% reduction in MSE and a 22% reduction in MAE.
  arXiv  Detail & Related papers  (2025-04-24T20:05:33Z)
• Large Continual Instruction Assistant [59.585544987096974]
  Continual Instruction Tuning (CIT) is adopted to instruct Large Models to follow human intent data by data. Existing update gradient would heavily destroy the performance on previous datasets during CIT process. We propose a general continual instruction tuning framework to address the challenge.
  arXiv  Detail & Related papers  (2024-10-08T11:24:59Z)
• FASTopic: Pretrained Transformer is a Fast, Adaptive, Stable, and Transferable Topic Model [76.509837704596]
  We propose FASTopic, a fast, adaptive, stable, and transferable topic model. We use Dual Semantic-relation Reconstruction (DSR) to model latent topics. We also propose Embedding Transport Plan (ETP) to regularize semantic relations as optimal transport plans.
  arXiv  Detail & Related papers  (2024-05-28T09:06:38Z)
• Fast TRAC: A Parameter-Free Optimizer for Lifelong Reinforcement Learning [6.388725318524439]
  Key challenge in lifelong reinforcement learning is the loss of plasticity. We present a parameter-free tuning for lifelong RL called TRAC. Experiments on Procgen, Atari, and Gym Control environments show TRAC works surprisingly well.
  arXiv  Detail & Related papers  (2024-05-26T17:38:44Z)
• Test-Time Model Adaptation with Only Forward Passes [68.11784295706995]
  Test-time adaptation has proven effective in adapting a given trained model to unseen test samples with potential distribution shifts. We propose a test-time Forward-Optimization Adaptation (FOA) method. FOA runs on quantized 8-bit ViT, outperforms gradient-based TENT on full-precision 32-bit ViT, and achieves an up to 24-fold memory reduction on ImageNet-C.
  arXiv  Detail & Related papers  (2024-04-02T05:34:33Z)
• Persistent Test-time Adaptation in Recurring Testing Scenarios [12.024233973321756]
  Current test-time adaptation (TTA) approaches aim to adapt a machine learning model to environments that change continuously. Yet, it is unclear whether TTA methods can maintain their adaptability over prolonged periods. We propose persistent TTA (PeTTA) which senses when the model is diverging towards collapse and adjusts the adaptation strategy.
  arXiv  Detail & Related papers  (2023-11-30T02:24:44Z)
• Fast-Slow Test-Time Adaptation for Online Vision-and-Language Navigation [67.18144414660681]
  We propose a Fast-Slow Test-Time Adaptation (FSTTA) approach for online Vision-and-Language Navigation (VLN) Our method obtains impressive performance gains on four popular benchmarks.
  arXiv  Detail & Related papers  (2023-11-22T07:47:39Z)
• Layer-wise Auto-Weighting for Non-Stationary Test-Time Adaptation [40.03897994619606]
  We introduce a layer-wise auto-weighting algorithm for continual and gradual TTA. We propose an exponential min-max scaler to make certain layers nearly frozen while mitigating outliers. Experiments on CIFAR-10C, CIFAR-100C, and ImageNet-C show our method outperforms conventional continual and gradual TTA approaches.
  arXiv  Detail & Related papers  (2023-11-10T03:54:40Z)
• Enhancing Plasticity for First Session Adaptation Continual Learning [20.62749699589017]
  We introduce Plasticity-Enhanced Test-Time Adaptation in Class-Incremental Learning (PLASTIC)<n>PLASTIC reinstates plasticity in CIL while preserving model stability.<n>It consistently outperforms both conventional and state-of-the-art PTM-based CIL approaches.
  arXiv  Detail & Related papers  (2023-10-17T13:06:39Z)
• Revisiting Plasticity in Visual Reinforcement Learning: Data, Modules and Training Stages [56.98243487769916]
  Plasticity, the ability of a neural network to evolve with new data, is crucial for high-performance and sample-efficient visual reinforcement learning. We propose Adaptive RR, which dynamically adjusts the replay ratio based on the critic's plasticity level.
  arXiv  Detail & Related papers  (2023-10-11T12:05:34Z)
• AR-TTA: A Simple Method for Real-World Continual Test-Time Adaptation [1.4530711901349282]
  We propose to validate test-time adaptation methods using datasets for autonomous driving, namely CLAD-C and SHIFT. We observe that current test-time adaptation methods struggle to effectively handle varying degrees of domain shift. We enhance the well-established self-training framework by incorporating a small memory buffer to increase model stability.
  arXiv  Detail & Related papers  (2023-09-18T19:34:23Z)
• PLASTIC: Improving Input and Label Plasticity for Sample Efficient Reinforcement Learning [54.409634256153154]
  In Reinforcement Learning (RL), enhancing sample efficiency is crucial. In principle, off-policy RL algorithms can improve sample efficiency by allowing multiple updates per environment interaction. Our study investigates the underlying causes of this phenomenon by dividing plasticity into two aspects.
  arXiv  Detail & Related papers  (2023-06-19T06:14:51Z)
• On Controller Tuning with Time-Varying Bayesian Optimization [74.57758188038375]
  We will use time-varying optimization (TVBO) to tune controllers online in changing environments using appropriate prior knowledge on the control objective and its changes. We propose a novel TVBO strategy using Uncertainty-Injection (UI), which incorporates the assumption of incremental and lasting changes. Our model outperforms the state-of-the-art method in TVBO, exhibiting reduced regret and fewer unstable parameter configurations.
  arXiv  Detail & Related papers  (2022-07-22T14:54:13Z)

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.