11 citations · 16 across the 18 of their papers we have counts for
13 papers · 1 filter
On the Limits of Representation-Based Test-Time Adaptation for Multimodal LLMs
Mikołaj Piórczyński, Wojciech Masarczyk, Damian Sójka +2
Vision-language models suffer significant accuracy drops under common corruptions. While test-time adaptation (TTA) successfully mitigates these shifts in CLIP-like contrastive mod…
Rethinking the Teacher-Student Framework for Test-Time Adaptation
Damian Sójka, Marc Masana, Bartłomiej Twardowski +1
Test-Time Adaptation (TTA) has recently emerged as a promising strategy that allows the adaptation of pre-trained models to changing data distributions at deployment time, without…
Subspace Optimization for Backpropagation-Free Continual Test-Time Adaptation
Damian Sójka, Sebastian Cygert, Marc Masana
We introduce PACE, a backpropagation-free continual test-time adaptation system that directly optimizes the affine parameters of normalization layers. Existing derivative-free appr…
Efficient Multi-Source Knowledge Transfer by Model Merging
Marcin Osial, Bartosz Wójcik, Bartosz Zieliński +1
While transfer learning is an effective strategy, it often overlooks the opportunity to leverage knowledge from numerous available models online. Addressing this multi-source trans…
No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces
Daniel Marczak, Simone Magistri, Sebastian Cygert +3
Model merging integrates the weights of multiple task-specific models into a single multi-task model. Despite recent interest in the problem, a significant performance gap between…
Task-recency bias strikes back: Adapting covariances in Exemplar-Free Class Incremental Learning
Grzegorz Rypeść, Sebastian Cygert, Tomasz Trzciński +1
Exemplar-Free Class Incremental Learning (EFCIL) tackles the problem of training a model on a sequence of tasks without access to past data. Existing state-of-the-art methods repre…
MagMax: Leveraging Model Merging for Seamless Continual Learning
Daniel Marczak, Bartłomiej Twardowski, Tomasz Trzciński +1
This paper introduces a continual learning approach named MagMax, which utilizes model merging to enable large pre-trained models to continuously learn from new data without forget…
Realistic Evaluation of Test-Time Adaptation Algorithms: Unsupervised Hyperparameter Selection
Sebastian Cygert, Damian Sójka, Tomasz Trzciński +1
Test-Time Adaptation (TTA) has recently emerged as a promising strategy for tackling the problem of machine learning model robustness under distribution shifts by adapting the mode…
Divide and not forget: Ensemble of selectively trained experts in Continual Learning
Grzegorz Rypeść, Sebastian Cygert, Valeriya Khan +3
Class-incremental learning is becoming more popular as it helps models widen their applicability while not forgetting what they already know. A trend in this area is to use a mixtu…
Revisiting Supervision for Continual Representation Learning
Daniel Marczak, Sebastian Cygert, Tomasz Trzciński +1
In the field of continual learning, models are designed to learn tasks one after the other. While most research has centered on supervised continual learning, there is a growing in…
Looking through the past: better knowledge retention for generative replay in continual learning
Valeriya Khan, Sebastian Cygert, Kamil Deja +2
In this work, we improve the generative replay in a continual learning setting to perform well on challenging scenarios. Current generative rehearsal methods are usually benchmarke…
Category Adaptation Meets Projected Distillation in Generalized Continual Category Discovery
Grzegorz Rypeść, Daniel Marczak, Sebastian Cygert +2
Generalized Continual Category Discovery (GCCD) tackles learning from sequentially arriving, partially labeled datasets while uncovering new categories. Traditional methods depend…
Adapt Your Teacher: Improving Knowledge Distillation for Exemplar-free Continual Learning
Filip Szatkowski, Mateusz Pyla, Marcin Przewięźlikowski +3
In this work, we investigate exemplar-free class incremental learning (CIL) with knowledge distillation (KD) as a regularization strategy, aiming to prevent forgetting. KD-based me…