相关论文: Continual Knowledge Consolidation LORA for Domain …

In the scenario of class-incremental learning (CIL), deep neural networks have to adapt their model parameters to non-stationary data distributions, e.g., the emergence of new classes over time. However, CIL models are challenged by the…

Continual instruction tuning (CIT) requires multimodal large language models (MLLMs) to adapt to a stream of tasks without forgetting prior capabilities. A common strategy is to isolate updates by routing inputs to different LoRA experts.…

One of the main challenges in reinforcement learning (RL) is that the agent has to make decisions that would influence the future performance without having complete knowledge of the environment. Dynamically adjusting the level of epistemic…

Previous continual learning setups for embodied intelligence focused on executing low-level actions based on human commands, neglecting the ability to learn high-level planning and multi-level knowledge. To address these issues, we propose…

Adaptability has been regarded as a central feature in the foundation models, enabling them to effectively acclimate to unseen downstream tasks. Parameter-efficient fine-tuning methods such as celebrated LoRA facilitate efficient adaptation…

Low-rank adaptation (LoRA) has emerged as a leading parameter-efficient fine-tuning technique for adapting large foundation models, yet it often locks adapters into suboptimal minima near their initialization. This hampers model…

Domain-Incremental Learning (DIL) enables vision models to adapt to changing conditions in real-world environments while maintaining the knowledge acquired from previous domains. Given privacy concerns and training time, Rehearsal-Free DIL…

The problem of continual learning has attracted rising attention in recent years. However, few works have questioned the commonly used learning setup, based on a task curriculum of random class. This differs significantly from human…

Low-rank Adaptation (LoRA) has demonstrated remarkable capabilities for task specific fine-tuning. However, in scenarios that involve multiple tasks, training a separate LoRA model for each one results in considerable inefficiency in terms…

Large language models (LLMs) exhibit remarkable capabilities in natural language processing but face catastrophic forgetting when learning new tasks, where adaptation to a new domain leads to a substantial decline in performance on previous…

Recently, pre-trained model and efficient parameter tuning have achieved remarkable success in natural language processing and high-level computer vision with the aid of masked modeling and prompt tuning. In low-level computer vision,…

Beyond general recognition tasks, specialized domains and fine-grained settings often encounter data scarcity, especially for tail classes. To obtain less biased and more reliable models under such scarcity, practitioners leverage diffusion…

Most existing works on continual learning (CL) focus on overcoming the catastrophic forgetting (CF) problem, with dynamic models and replay methods performing exceptionally well. However, since current works tend to assume exclusivity or…

The ability to learn in dynamic, nonstationary environments without forgetting previous knowledge, also known as Continual Learning (CL), is a key enabler for scalable and trustworthy deployments of adaptive solutions. While the importance…

LoRA has become a universal Parameter-Efficient Fine-Tuning (PEFT) technique that equips Large Language Models (LLMs) to adapt quickly to new tasks. However, when these models are scaled up, even the latest LoRA variants still introduce…

Continual Learning (CL) enables machine learning models to learn from continuously shifting new training data in absence of data from old tasks. Recently, pretrained vision transformers combined with prompt tuning have shown promise for…

LoRA-based continual learning represents a promising avenue for leveraging pre-trained models in downstream continual learning tasks. Recent studies have shown that orthogonal LoRA tuning effectively mitigates forgetting. However, this work…

We propose Deep Optimistic Linear Support Learning (DOL) to solve high-dimensional multi-objective decision problems where the relative importances of the objectives are not known a priori. Using features from the high-dimensional inputs,…

We address the question of how to successively add new knowledge to an LLM whilst retaining previously-added knowledge. We consider two settings, semi-cooperative and fully-cooperative. Overall, LoRA performs better in most cases than…

Detectors trained with massive labeled data often exhibit dramatic performance degradation in some particular scenarios with data distribution gap. To alleviate this problem of domain shift, conventional wisdom typically concentrates solely…