25 accepted papers
Recent advances in large language models (LLMs) test-time computing have introduced the capability to perform intermediate chain-of-thought (CoT) reasoning (thinking) before generating answers. While increasing the thinking budget yields smooth performance improvements at inference time, the relatio…
Multimodal change detection (MCD) has important applications in disaster assessment, but the nonlinear distortion of features and spatial misalignment caused by sensor imaging differences make it difficult to obtain changes through direct comparison. To overcome the above problems, this study aims t
In medical image segmentation tasks, the domain gap caused by the difference in data collection between training and testing data seriously hinders the deployment of pre-trained models in clinical practice. Continual Test-Time Adaptation (CTTA) aims to enable pre-trained models to adapt to continuou
Existing semi-supervised learning methods typically mitigate the impact of unreliable predictions by suppressing low-confidence regions. However, these methods fail to explore which regions hold higher learning value and how to design adaptive learning strategies for these regions. To address these…
Due to the inherent inductive bias of operations, convolutional neural networks (CNN) cannot model global information of remote sensing (RS) images. In contrast, Transformer-based methods can establish long-range dependencies of images through self-attention (SA) mechanism, but it faces the challeng…
Diffusion transformers have been widely adopted for text-to-image synthesis. While scaling these models up to billions of parameters shows promise, the effectiveness of scaling beyond current sizes remains underexplored and challenging. By explicitly exploiting the computational heterogeneity of ima…
With the rapid scaling of large language models (LLMs), structured pruning has become a widely used technique to learn efficient, smaller models from larger ones, delivering superior performance compared to training similarly sized models from scratch. In this paper, we move beyond the traditional s…
"In this work, we discuss building performant Multimodal Large Language Models (MLLMs). In particular, we study the importance of various architecture components and data choices. Through careful and comprehensive ablations of the image encoder, the vision language connector, and various pre-trainin…
The hybrid architecture of convolutional neural networks (CNNs) and Transformer are very popular for medical image segmentation. However, it suffers from two challenges. First, although a CNNs branch can capture the local image features using vanilla convolution, it cannot achieve adaptive feature l…
Many natural language processing tasks benefit from long inputs, but processing long documents with Transformers is expensive -- not only due to quadratic attention complexity but also from applying feedforward and projection layers to every token. However, not all tokens are equally important, espe…
We propose Conditional Adapter (CoDA), a parameter-efficient transfer learning method that also improves inference efficiency. CoDA generalizes beyond standard adapter approaches to enable a new way of balancing speed and accuracy using conditional computation. Starting with an existing dense pretra…
The popular networks for change detection (CD) in very-high-resolution (VHR) remote sensing (RS) images usually suffer from two problems. First, it is difficult for these networks to model simultaneously the local and global features of changed targets, which leads to the limited feature representat…
Multi-vector retrieval models such as ColBERT [Khattab et al., 2020] allow token-level interactions between queries and documents, and hence achieve state of the art on many information retrieval benchmarks. However, their non-linear scoring function cannot be scaled to millions of documents, necess…
The popular convolutional neural networks (CNNs) have been successfully used in very high-resolution remote sensing image semantic segmentation. However, these networks often suffer from performance limitations. First, although deeper networks usually provide better feature representation, they may…
Sparsely-activated Mixture-of-experts (MoE) models allow the number of parameters to greatly increase while keeping the amount of computation for a given token or a given sample unchanged. However, a poor expert routing strategy (e.g. one resulting in load imbalance) can cause certain experts to be…
The Transformer architecture has been well adopted as a dominant architecture in most sequence transduction tasks including automatic speech recognition (ASR), since its attention mechanism excels in capturing long-range dependencies. While models built solely upon attention can be better paralleliz…
Recent work has improved language models (LMs) remarkably by equipping them with a non-parametric memory component. However, most existing approaches only introduce mem-ories at testing time or represent them using a separately trained encoder, resulting in suboptimal training of the language model.…
The popular deep convolutional networks used for image super-resolution (SR) reconstruction often increase the network depth and employ attention mechanism to improve image reconstruction effect. However, these networks suffer from two problems. The first is the deeper network easily causes higher c…
We present a method for generating comparative summaries that highlight similarities and contradictions in input documents. The key challenge in creating such summaries is the lack of large parallel training data required for training typical summarization systems. To this end, we introduce a hybrid…
We introduce Nutri-bullets, a multi-document summarization task for health and nutrition. First, we present two datasets of food and health summaries from multiple scientific studies. Furthermore, we propose a novel extract-compose model to solve the problem in the regime of limited parallel data. W…
Large language models have become increasingly difficult to train because of the growing computation time and cost. In this work, we present SRU++, a highly-efficient architecture that combines fast recurrence and attention for sequence modeling. SRU++ exhibits strong modeling capacity and training…
The V-Net based 3D fully convolutional neural networks have been widely used in liver volumetric data segmentation. However, due to the large number of parameters of these networks, 3D FCNs suffer from high computational cost and GPU memory usage. To address these issues, we design a lightweight V-N…
In this paper, we propose a novel approach based on a symmetric fully convolutional network within pyramid pooling (FCN-PP) for landslide mapping (LM). The proposed approach has three advantages. Firstly, this approach is automatic and insensitive to noise because multivariate morphological reconstr…
The design of neural architectures for structured objects is typically guided by experimental insights rather than a formal process. In this work, we appeal to kernels over combinatorial structures, such as sequences and graphs, to derive appropriate neural operations. We introduce a class of deep r…
This paper focuses on style transfer on the basis of non-parallel text. This is an instance of a broad family of problems including machine translation, decipherment, and sentiment modification. The key challenge is to separate the content from other aspects such as style. We assume a shared latent…