62 accepted papers
As an essential component of fine-tuning, warm-up plays a crucial role in promoting stability and generalization. Many studies have examined its underlying mechanisms from different aspects. However, most of the studies focus on incorporating these insights into optimizers to reduce the reliance on
Continuum robots exhibit exceptional flexibility and multi-degree-of-freedom maneuverability, offering significant advantages for navigating confined luminal spaces. However, rapid simulation of their contact-influenced behavior remains challenging. This paper presents RLFEM, an innovative finite el
Autoregressive large language models (LLMs) deliver strong performance but require inherently sequential decoding, leading to high inference latency and poor GPU utilization. Speculative decoding mitigates this bottleneck by using a fast draft model whose outputs are verified in parallel by the targ…
Video large language models (Video-LLMs) have demonstrated strong capabilities in video understanding tasks. However, their practical deployment is still hindered by the inefficiency introduced by processing massive amounts of visual tokens. Although recent approaches achieve extremely low token ret
Stickers are widely used in online communication to convey emotions and implicit intentions. The Sticker Response Selection (SRS) task aims to select the most contextually appropriate sticker based on the dialogue. However, existing methods typically rely on semantic matching and model emotional and
Recent advancements in text-to-image (T2I) generation have led to the emergence of highly expressive models such as diffusion transformers (DiTs), exemplified by FLUX. However, their massive parameter sizes lead to slow inference, high memory usage, and poor deployability. Existing acceleration meth
Embodied navigation is a fundamental capability for intelligent agents, yet remains challenging in partially observable environments where navigation instructions can be difficult to interpret. However, existing tasks only provide unimodal instructions, which are ambiguous in complex multimodal envi
Multimodal large language models (MLLMs) are plagued by exorbitant inference costs attributable to the profusion of visual tokens within the vision encoder. The redundant visual tokens engenders a substantial computational load and key-value (KV) cache footprint bottleneck. Existing approaches focus
The prohibitive cost of 3D attention hinders high-quality video generation with diffusion models. Existing sparse attention methods either lack content adaptivity (static) or incur excessive overhead from per-step recalculation (dynamic). Our work challenges the necessity of this trade-off, based on
Vision Transformers (ViTs) have achieved strong performance in video action recognition, but their high computational cost limits their practicality. Lightweight CNNs are more efficient but suffer from accuracy gaps. Cross-Architecture Knowledge Distillation (CAKD) addresses this by transferring kno
Electrocardiogram (ECG) representation learning via ECG-report alignment is often hindered by the inherent structural and statistical divergence between signals and natural language. Existing methods struggle to bridge this gap with simple contrastive objectives, but struggle with distribution depen…
The deployment of Large Language Models is constrained by the memory and bandwidth demands of static weights and dynamic Key-Value cache. SVD-based compression provides a hardware-friendly solution to reduce these costs. However, existing methods suffer from two key limitations: some are suboptimal …
Optimizers are critical for training deep neural networks. Existing training processes rely on a single static optimizer (e.g., SGD) or a simple hybrid of two optimizers, which miss the opportunity to exploit evolving dynamics in different training states, degrading model quality and convergence. In…
Continuum robots offer high flexibility and multiple degrees of freedom, making them ideal for navigating narrow lumens. However, accurately modeling their behavior under large deformations and frequent environmental contacts remains challenging. Current methods for solving the deformation of these
Backdoor attacks pose a significant threat to deep neural networks (DNNs), as attackers can inject a backdoor by tampering with only a few samples. The variety of backdoor attacks makes comprehensive defense extremely challenging. Previous defenses typically assume that backdoor samples are out-of-d…
Electrocardiogram (ECG) is widely used to diagnose cardiac conditions via deep learning methods. Although existing self-supervised learning (SSL) methods have achieved great performance in learning representation for ECG-based cardiac conditions classification, the clinical semantics can not be effe
Natural Language to SQL (NL2SQL) provides a new model-centric paradigm that simplifies database access for non-technical users by converting natural language queries into SQL commands. Recent advancements, particularly those integrating Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT)
A critical prerequisite for achieving generalizable robot control is the availability of a large-scale robot training dataset. Due to the expense of collecting realistic robotic data, recent studies explored simulating and recording robot skills in virtual environments. While simulated data can be g…
Self-supervised learning (SSL) has garnered increasing attention in electrocardiogram (ECG) analysis for its effectiveness in resource-limited settings. Existing state-of-the-art SSL methods rely on time-frequency detail reconstruction, but due to the inherent redundancy of ECG signals and individua
Text summarization task extracts salient information from a large amount of text for productivity enhancement. However, most existing methods heavily rely on training models from ample and centrally stored data which is infeasible to collect in practice, due to privacy concerns and data scarcity nat…
Federated Learning (FL) is a prominent distributed learning approach that addresses two major challenges: statistical heterogeneity (i.e., non-identically distributed data) and system heterogeneity (i.e., variability in the communication and computation on each client). As FL is commonly applied in
Graphical User Interface (GUI) agents, powered by Large Foundation Models, have emerged as a transformative approach to automating human-computer interaction. These agents autonomously interact with digital systems via GUIs, emulating human actions such as clicking, typing, and navigating visual ele…
Knowledge distillation aims to transfer knowledge from a large teacher model to a compact student counterpart, often coming with a significant performance gap between them. Interestingly, we find that a too-large performance gap can hamper the training process. To alleviate this, we propose a **Gap…
We study the problem of classification with selectively labeled data, whose distribution may differ from the full population due to historical decision-making. We exploit the fact that in many applications historical decisions were made by multiple decision-makers, each with different decision rules…
Large Language Models (LLMs) have become more prevalent in long-context applications such as interactive chatbots, document analysis, and agent workflows, but it is challenging to serve long-context requests with low latency and high throughput. Speculative decoding (SD) is a widely used technique t…
We introduce LLaVA-Reward, an efficient reward model designed to automatically evaluate text-to-image (T2I) generations across multiple perspectives, leveraging pretrained multimodal large language models (MLLMs). Existing MLLM-based approaches require instruction-following data for supervised fine-…
A core data-centric learning challenge is the identification of training samples that are detrimental to model performance. Influence functions serve as a prominent tool for this task and offer a robust framework for assessing training data influence on model predictions. Despite their widespread us…
Laparoscopic ultrasound (LUS) serves as a critical technology in intraoperative surgeries, particularly for guiding complex procedures in liver diseases. However, the development of robotic systems for LUS examination remains hindered by challenges such as high costs and the absence of force feedbac
Diffusion models are prominent in image generation for producing detailed and realistic images from Gaussian noises. However, they often encounter instability issues in image restoration tasks, e.g., super-resolution. Existing methods typically rely on multiple runs to find an initial noise that pro…
Multimodal large language models (MLLMs) enhance their perceptual capabilities by integrating visual and textual information. However, processing the massive number of visual tokens incurs a significant computational cost. Existing analysis of the MLLM attention mechanisms remains shallow, leading t…
Multimodal large language models (MLLMs) have recently shown great progress in text-rich image understanding, yet they still struggle with complex, multi-page visually-rich documents. Traditional methods using document parsers for retrieval-augmented generation suffer from performance and efficiency…
Contrastive Learning (CL) plays a crucial role in molecular representation learning, enabling unsupervised learning from large scale unlabeled molecule datasets. It has inspired various applications in molecular property prediction and drug design. However, existing molecular representation learning…
Offline model-based reinforcement learning (MBRL) enhances data efficiency by utilizing pre-collected datasets to learn models and policies, especially in scenarios where exploration is costly or infeasible. Nevertheless, its performance often suffers from the objective mismatch between model and po…
Bronchoscopy plays a significant role in the early diagnosis and treatment of lung diseases. This process demands physicians to maneuver the flexible endoscope for reaching distal lesions, particularly requiring substantial expertise when examining the airways of the upper lung lobe. With the develo…
Transorbital Neuro Surgery (TNS) offers a novel treatment towards the lesion inside skull pursuing minimal invasiveness. Most conventional TNS tools are rigid and straight, limiting the dexterity and accessibility in passing a small port. Bendable and steerable surgical tools provides an alternative…
Real-time 6 DOF localization of bronchoscopes is crucial for enhancing intervention quality. However, current vision-based technologies struggle to balance between generalization to unseen data and computational speed. In this study, we propose a Depth-based Dual-Loop framework for real-time Visuall
Traditional rigid endoscopes have challenges in flexibly treating tumors located deep in the brain, and low operability and fixed viewing angles limit its development. This study introduces a novel dual-segment flexible robotic endoscope MicroNeuro, designed to perform biopsies with dexterous surgic…
Accurate evaluation of financial question answering (QA) systems necessitates a comprehensive dataset encompassing diverse question types and contexts. However, current financial QA datasets lack scope diversity and question complexity. This work introduces FinTextQA, a novel dataset for long-form q…
Robot-assisted minimally invasive surgery (MIS) using flexible endoscopy has emerged as a groundbreaking technology for improving traditional surgical approaches. However, a major challenge in advancing this technology is the lack of shape sensing, which leads to inaccurate navigation and control of
In Graph Convolutional Networks (GCNs), the aggregation of node features in graph convolutional learning is typically guided solely by the topology of the graphs. However, both network topology and node features provide unique and valuable information. Relying solely on topology cannot yield entirel…
Large multimodal language models have shown remarkable proficiency in understanding and editing images. However a majority of these visually-tuned models struggle to comprehend the textual content embedded in images primarily due to the limitation of training data. In this work we introduce TRINS: a…
Automatic generation of graphical layouts is crucial for many real-world applications, including designing posters, flyers, advertisements, and graphical user interfaces. Given the incredible ability of Large language models (LLMs) in both natural language understanding and generation, we believe th…
Bone marrow smear cytology involves observing and analyzing the morphological features of bone marrow cells, and identifying regions of interest (ROI) where the cells are morphologically clear and evenly distributed is a crucial part of this process. However, existing deep learning methods for selec…
Controllable layout generation refers to the process of creating a plausible visual arrangement of elements within a graphic design (*e.g.*, document and web designs) with constraints representing design intentions. Although recent diffusion-based models have achieved state-of-the-art FID scores, th…
Deep networks have achieved great success in image rescaling (IR) task that seeks to learn the optimal downscaled representations, i.e., low-resolution (LR) images, to reconstruct the original high-resolution (HR) images. Compared with super-resolution methods that consider a fixed downscaling schem…
The roadside video surveillance signal can help people achieve vehicle tracking and trajectory generation. Using these trajectories can learn the future motion of vehicles. Existing prediction methods can not analyze the interaction between vehicles well. To this end, we design a dynamic vehicle gra…
Learning to navigate to an image-specified goal is an important but challenging task for autonomous systems like household robots. The agent is required to well understand and reason the location of the navigation goal from a picture shot in the goal position. Existing methods try to solve this prob…
Current imaging of the artery relies primarily on computed tomography angiography (CTA), which requires contrast injections and exposure to radiation. In this paper, we present a method for fully autonomous artery 3D image acquisition using a linear ultrasound (US) probe and a 6 DoFs robot arm with…
The accurate prediction of the neighboring vehicles' trajectories affects the security of autonomous driving vehicles. However, it is challenging for existing methods to anticipating the trajectories of vehicles in the vicinity due to the uncertainty of driving behaviors and the complex interaction…
Learning from noisy labels is an important and long-standing problem in machine learning for real applications. One of the main research lines focuses on learning a label corrector to purify potential noisy labels. However, these methods typically rely on strict assumptions and are limited to certai…
Deep neural network (DNN) shows impressive performance on many tasks but they usually suffer from adversarial examples with human eyes invisible slight perturbation. Such examples can not be distinguished by human but can mislead DNN classifiers leading to its important role in DNN attack and defens…
We present a novel network, PI-Net, for the fusion between point clouds and images in this letter. Most existing fusion methods project point clouds into pseudo images and then fuse the pseudo and RGB images with 2D CNNs. To get rid of structuring the pseudo images as the preprocessing, we propose a
For multihypothesis-based compressed video sensing schemes, the low accuracy of weight prediction and degradation of recovery quality for high-motion videos are open challenges. To solve this problem, this paper proposes a new multihypothesis prediction scheme. To efficiently get high-quality hypoth…
Neural architecture search (NAS) has become an important approach to automatically find effective architectures. To cover all possible good architectures, we need to search in an extremely large search space with billions of candidate architectures. More critically, given a large search space, we ma…
Deep neural networks have exhibited promising performance in image super-resolution (SR) by learning a nonlinear mapping function from low-resolution (LR) images to high-resolution (HR) images. However, there are two underlying limitations to existing SR methods. First, learning the mapping function…
End-to-end Task-oriented spoken dialog systems typically require modeling two types of inputs, namely, the dialog history which is a sequence of utterances and the knowledge base (KB) associated with the dialog history. While modeling these inputs, current state-of-the-art models typically ignore th…
Learning robotic grasps from visual observations is a promising yet challenging task. Recent research shows its great potential by preparing and learning from large-scale synthetic datasets. For the popular, 6 degree-of-freedom (6-DOF) grasp setting of parallel-jaw gripper, most of existing methods…
Maximum A Posterior (MAP) acts as one of the most popular modeling scheme in image restoration and is usually reduced to a separable optimization model. Unfortunately, it is challenging to establish exact regularization term and the model with complex priors is hard to optimize. In additionally, it…
Weld pool penetration is a major indicator of weld integrity but it is not directly measurable. The key to solving this issue is to find or create observable phenomena that can be used to derive the penetration. An innovative method has recently been proposed/developed to derive the penetration from
Designing effective architectures is one of the key factors behind the success of deep neural networks. Existing deep architectures are either manually designed or automatically searched by some Neural Architecture Search (NAS) methods. However, even a well-searched architecture may still contain ma…
This paper presents a 2-point algorithm for relative pose estimation of a monocular camera mounted on a ground vehicle. Specifically, the geometric model is developed by combining two-view geometry with planar motion constraint. Based on the proposed model, the relative camera pose information can b…
This paper presents a trifocal tensor based approach for the visual trajectory tracking task of a nonholonomic mobile robot, which is equipped with a roughly installed monocular camera. A set of pre-recorded images are used to express the desired trajectory, and the robot is regulated to track the d…