29 accepted papers
We study offline change point localization and inference in dynamic multilayer random dot product graphs (D-MRDPGs), where at each time point, a multilayer network is observed with shared node latent positions and time-varying, layer-specific connectivity patterns. We propose a novel two-stage algor…
Novel view synthesis (NVS) under large view deviations remains an underexplored challenge for 3D Gaussian Splatting (3DGS). In urban scenes with limited training coverage, models often fail to maintain geometric consistency when extrapolating to unseen viewpoints, resulting in severe distortions and
Lifelong Model Editing aims to continuously update evolving facts in Large Language Models while preserving unrelated knowledge and general capabilities, yet it remains plagued by catastrophic forgetting and model collapse. Empirically, we find that the few recent editors resilient over long horizon…
As one of the effective closed-loop control methods, visual servoing control methods are widely applied to continuum robots. However, existing visual servoing control methods mostly focus on accurate control of the robot’s end-effector, with less consideration given to the robot’s shape. In this wor
For objects with complex topological and geometrical features, stochastic topological grasping can be executed without the necessity for feedback or precise planning. However, this grasping method has two significant limitations. First, the technique's effectiveness is reduced when interacting with
In nature, prehensile tails serve as versatile and essential appendages for animals, facilitating both grasping and enhanced mobility. Although existing robotic tails effectively contribute to mobility across a range of behaviors, they lack versatile object-grasping capabilities. Inspired by these b
Diffusion models have achieved significant progress in the task of image animation due to their powerful generative capabilities. However, preserving appearance consistency to the static input image, and avoiding abrupt motion change in the generated animation, remains challenging. In this paper, we…
Ensuring both mobility and stability during colonoscopy is crucial for enhancing procedural efficiency and reducing the risk of complications. Traditional colonoscopy robots face challenges due to the fixed diameter of the colonoscope and the variability in colon anatomy. To address this, we propose
Age-related sarcopenia weakens balance in older adults, highlighting the need for effective assistive robots. How-ever, existing assistive technologies often suffer from mechanical, kinematic, and control incompatibilities with the human body. Here, we present a wearable assistive robotic system tha
Graph transfer learning, especially in unsupervised domain adaptation, aims to transfer knowledge from a label-abundant source graph to an unlabeled target graph. However, most existing approaches overlook the common issue of label imbalance in the source domain, typically assuming a balanced label…
Novel view synthesis is a critical task in autonomous driving. Although 3D Gaussian Splatting (3D-GS) has shown success in generating novel views, it faces challenges in maintaining high-quality rendering when viewpoints deviate significantly from the training set. This difficulty primarily stems fr…
Graph domain adaptation is a key subfield of graph transfer learning that aims to bridge domain gaps by transferring knowledge from a label-rich source graph to an unlabeled target graph. However, most existing methods assume balanced labels in the source graph, which often fails in practice and lea
Flexible endoscopes are commonly adopted in narrow and confined anatomical cavities due to their higher reachability and dexterity. However, prolonged and unintuitive manipulation of these endoscopes leads to an increased workload on surgeons and risks of collision. To address these challenges, this
This paper introduces InternVid, a large-scale video-centric multimodal dataset that enables learning powerful and transferable video-text representations for multimodal understanding and generation. InternVid contains over 7 million videos lasting nearly 760K hours, yielding 234M video clips accomp…
Large language models (LLMs), although having revolutionized many fields, still suffer from the challenging extrapolation problem, where the inference ability of LLMs sharply declines beyond their max training lengths. In this work, we conduct a theoretical analysis to better understand why No Posit…
Recently video generation has achieved substantial progress with realistic results. Nevertheless, existing AI-generated videos are usually very short clips ("shot-level'') depicting a single scene. To deliver a coherent long video ("story-level''), it is desirable to have creative transition and pre…
Mode collapse remains the primary unresolved challenge within generative adversarial networks (GANs). In this work, we introduce an innovative approach that supplements the discriminator by additionally enforcing the similarity between the generated and real distributions. We implement a one-sample…
Few-sample compression aims to compress a big redundant model into a small compact one with only few samples. If we fine-tune models with these limited few samples directly, models will be vulnerable to overfit and learn almost nothing. Hence, previous methods optimize the compressed model layer-by-…
Existing robotic endoscopes for laparoscopic surgery, predominantly rigid or limited in dexterity, occupy a large motion space1, The large occupied motion space necessitates large incisions and reduces the motion space for surgeons to simultaneously operate other surgical instruments. Meanwhile, sur…
Recent deep clustering algorithms take advantage of self-supervised learning and self-training techniques to map the original data into a latent space, where the data embedding and clustering assignment can be jointly optimized. However, as many recent datasets are enormous and noisy, getting a clea…
The beneficial aspects of fully-actuated multirotor UAVs, provided by nonparallel rotor configuration, are increasingly being recognized and utilized to great benefit in high-precision applications. Full six-degree-of-freedom force control, higher control bandwidth and improved disturbance rejection…
Everyday contact-rich tasks, such as peeling, cleaning, and writing, demand multimodal perception for effective and precise task execution. However, these present a novel challenge to robots as they lack the ability to combine these multimodal stimuli for performing contact-rich tasks. Learning-base…
A flexible endoscope introduces more dexterity to the image capturing in endoscopic surgery. However, manual control or automatic control based on instrument tracking does not handle the misorientation between the endoscopic video and the surgeon. We propose an automatic flexible endoscope control m…
Inspection for structural properties (surface stiffness and coefficient of restitution) is crucial for understanding and performing aerial manipulations in unknown environments, with little to no prior knowledge on their state. Inspection-on-the-fly is the uncanny ability of humans to infer states d…
Endoscopes play an important role in minimally invasive surgery (MIS). Due to the advantages of less occupied motion space and enhanced safety, flexible endoscopes are drawing more and more attention. However, the structure of the flexible section makes it difficult for surgeons to manually rotate a
Automation in robotic surgery has become an increasingly attractive topic. Although full automation remains fictional, task autonomy and conditional autonomy are highly achievable. Apart from the performance of task fulfillment, one major concern in robotic surgery is safety. In this paper, we prese
Visual loop closure detection, which can be considered as an image retrieval task, is an important problem in SLAM (Simultaneous Localization and Mapping) systems. The frequently used bag-of-words (BoW) models can achieve high precision and moderate recall. However, the requirement for lower time co…
Automation in surgery is becoming more and more active in recent years. From the editorial of Science Robotics in 2017, there are six levels of autonomy and the existing tele-operation lies in level 0 (no autonomy). Although full autonomy would be fictional in the current moment, task/shared autonom
Object tracking is an important capability for robots tasked with interacting with humans and the environment, and it enables robots to manipulate objects. In object tracking, selecting samples to learn a robust and efficient appearance model is a challenging task. Model learning determines both the…