Related papers: MOMA: Masked Orthogonal Matrix Alignment for Zero-…

Model merging has emerged as an efficient method to combine multiple single-task fine-tuned models. The merged model can enjoy multi-task capabilities without expensive training. While promising, merging into a single model often suffers…

Abstaining classifiers have been widely used in cost-sensitive applications to avoid ambiguous classification and reduce the cost of misclassification. Previous abstaining classification models rely on cost information, such as a cost…

In this paper, we reveal that most current efficient multimodal fine-tuning methods are hindered by a key limitation: they are directly borrowed from LLMs, often neglecting the intrinsic differences of multimodal scenarios and even…

Applying pre-trained models to assist point cloud understanding has recently become a mainstream paradigm in 3D perception. However, existing application strategies are straightforward, utilizing only the final output of the pre-trained…

Model merging combines knowledge from task-specific models into a unified multi-task model to avoid joint training on all task data. However, current methods face challenges due to representation bias, which can interfere with tasks…

Masked Image Modeling (MIM) is a technique in self-supervised learning that focuses on acquiring detailed visual representations from unlabeled images by estimating the missing pixels in randomly masked sections. It has proven to be a…

Model merging is an effective post-training strategy for composing capabilities in large language models without joint retraining. We study this in the supervised fine-tuning (SFT) stage, where multiple capability-based SFT checkpoints --…

Learning-enabled control systems increasingly rely on multiple sensing modalities (e.g., vision, audio, language, etc.) for perception and decision support. A key challenge is that multi-modal sensor training dynamics are often imbalanced:…

Manifold alignment (MA) involves a set of techniques for learning shared representations across domains, yet many traditional MA methods are incapable of performing out-of-sample extension, limiting their real-world applicability. We…

Model merging combines independently fine-tuned checkpoints without joint multi-task training. In the era of foundation-model, fine-tuning with Low-Rank Adaptation (LoRA) is prevalent, making LoRA merging a promising target. Existing…

Vision-and-Language Pre-training (VLP) improves model performance for downstream tasks that require image and text inputs. Current VLP approaches differ on (i) model architecture (especially image embedders), (ii) loss functions, and (iii)…

We propose a method that morphs high-orger meshes such that their boundaries and interfaces coincide/align with implicitly defined geometries. Our focus is particularly on the case when the target surface is prescribed as the zero…

Multi-modal MRIs are widely used in neuroimaging applications since different MR sequences provide complementary information about brain structures. Recent works have suggested that multi-modal deep learning analysis can benefit from…

Fine-tuning pre-trained models with custom data leads to numerous expert models on specific tasks. Merging models into one universal model to empower multi-task ability refraining from data leakage has gained popularity. With the expansion…

Fine-tuning Large Language Models (LLMs) and storing them for each downstream task or domain is impractical because of the massive model size (e.g., 350GB in GPT-3). Current literature, such as LoRA, showcases the potential of low-rank…

The widespread availability of fine-tuned LoRA modules for open pre-trained models has led to an interest in methods that can adaptively merge LoRAs to improve performance. These methods typically include some way of selecting LoRAs from a…

The application of multiple-input multiple-output (MIMO) techniques to non-orthogonal multiple access (NOMA) systems is important to enhance the performance gains of NOMA. In this paper, a novel MIMO-NOMA framework for downlink and uplink…

Training large language models (LLMs) relies almost exclusively on dense adaptive optimizers with increasingly sophisticated preconditioners. We challenge this by showing that randomly masking parameter updates can be highly effective, with…

This paper presents the Multi-Objective Ant Nesting Algorithm (MOANA), a novel extension of the Ant Nesting Algorithm (ANA), specifically designed to address multi-objective optimization problems (MOPs). MOANA incorporates adaptive…

Training a unified model to take multiple targets into account is a trend towards artificial general intelligence. However, how to efficiently mitigate the training conflicts among heterogeneous data collected from different domains or…