77 accepted papers
Time series forecasting under distribution shift remains challenging, as existing deep learning models often rely on local statistical normalization (e.g., mean and variance) that fails to capture global distribution shift. Methods like RevIN and its variants attempt to decouple distribution and pat
Recent 3D Large-Language Models (3D-LLMs) claim to understand 3D worlds, especially spatial relationships among objects. Yet, we find that simply fine-tuning a language model on text-only question-answer pairs can perform comparably or even surpass these methods on the SQA3D benchmark without using…
Compositional diffusion planners enable robotic decision-making beyond the horizon of training trajectories. Yet, current approaches often rely on the heuristic stitching of local predictions. We demonstrate that this induces a non-conservative vector field that does not mathematically correspond to…
Large language models (LLMs) have demonstrated remarkable performance due to their large parameter counts and extensive training data. However, their scale leads to significant memory bottlenecks during training, especially when using memory-intensive optimizers like Adam. Existing memory-efficient …
In modern software development, particularly in emerging ``vibe coding'' paradigms, project implementation increasingly begins with visual interactions between users and AI coding assistants, where system architectures are communicated through visual designs before coding. This visual-first approach…
Academic posters are vital for scholarly communication, yet their manual creation is time-consuming. However, automated academic poster generation faces significant challenges in preserving intricate scientific details and achieving effective visual-textual integration. Existing approaches often str…
The impressive performance of large language models (LLMs) arises from their massive scale and heterogeneous module composition. However, this structural heterogeneity poses significant optimization challenges. While adaptive optimizers such as Adam(W) provide per-parameter adaptivity, they do not e
Achieving efficient and uniform coverage in obstacle-laden unknown environments is essential for au- tonomous robots in cleaning, inspection and agricultural op- erations. Unlike most existing approaches that prioritize path length and time optimality, we propose the SHIFT planner framework, which i…
Stochastic gradient methods are central to large-scale learning, yet their generalization theory typically relies on independent sampling assumptions. In many practical applications, data are generated by Markov chains and learning is performed in a decentralized manner, which introduces significant
Shared control is the most commonly used physical human-robot interaction (pHRI) in robot-assisted orthopedic surgery. However, challenges persist in tasks such as robotassisted knee arthroplasty, particularly in terms of motion delays and inadequate three-dimensional (3D) constraints. In this lette
In this letter, an active-passive fusion rehabilitation training method for the ankle fracture rehabilitation robot after surgery is developed. A subject-independent continuous estimation model of ankle torque is proposed based on surface electromyography (sEMG), an online adaptive algorithm for pas
Unmanned ground vehicles operating in complex environments must adaptively adjust to modeling uncertainties and external disturbances to perform tasks such as wall following and obstacle avoidance. This paper introduces an adaptive control approach based on spiking neural networks for wall fitting a
The emergence of Vision-Language Models (VLMs) represents a significant advancement in integrating computer vision with Large Language Models (LLMs) to generate detailed text descriptions from visual inputs. Despite their growing importance, the security of VLMs, particularly against backdoor attack…
Decentralized learning, which facilitates joint model training across geographically scattered devices, has gained significant attention in the field of signal and information processing in recent years. While the optimization errors of decentralized learning algorithms have been extensively studied…
Weight decay is a widely used technique in training machine learning models, known to empirically enhance the generalization of Stochastic Gradient Descent (SGD). While intuitively weight decay allows SGD to train a regularized model rather than the original one, there is limited theoretical underst…
Code large language models (LLMs) have shown remarkable advances in code understanding, completion, and generation tasks. Programming benchmarks, comprised of a selection of code challenges and corresponding test cases, serve as a standard to evaluate the capability of different LLMs in such tasks.…
For inductive link prediction in knowledge graphs, we address the problem by considering bridging links (i.e., links connecting discrete graphs). Although current research overcomes the traditional graph topological constraints, it tends to ignore the dynamic interactions between relations and entit…
Large Language Model (LLM)-based multi-agent systems show promise for automating real-world tasks but struggle to transfer across domains due to their domain-specific nature. Current approaches face two critical shortcomings: they require complete architectural redesign and full retraining of all co…
Meta-learning has achieved significant advancements, with generalization emerging as a key metric for evaluating meta-learning algorithms. While recent studies have mainly focused on training strategies, data-split methods, and tightening generalization bounds, they often ignore the impact of inner-…
We present Rectified Point Flow, a unified parameterization that formulates pairwise point cloud registration and multi-part shape assembly as a single conditional generative problem. Given unposed point clouds, our method learns a continuous point-wise velocity field that transports noisy points to…
Preference alignment in Large Language Models (LLMs) has significantly improved their ability to adhere to human instructions and intentions. However, existing direct alignment algorithms primarily focus on relative preferences and often overlook the qualitative aspects of responses, despite having…
Spatio-temporal trajectory representation learning plays a crucial role in various urban applications such as transportation systems, urban planning, and environmental monitoring. Existing methods can be divided into single-view and multi-view approaches, with the latter offering richer representati…
Sharpness-Aware Minimization (SAM) has proven highly effective in improving model generalization in machine learning tasks. However, SAM employs a fixed hyperparameter associated with the regularization to characterize the sharpness of the model. Despite its success, research on adaptive regularizat…
Pruning is a widely used technique for compressing large neural networks that eliminates weights that have minimal impact on the model's performance. Current pruning methods, exemplified by magnitude pruning, assign an importance score to each weight based on its magnitude and remove weights with sc…
Coupling-Tiltable Unmanned Aerial-Aquatic Vehicles (UAAVs) have gained increasing importance, yet lack comprehensive analysis and suitable controllers. This paper analyzes the underwater motion characteristics of a self-designed UAAV, Mirs-Alioth, and designs a controller for it. The effectiveness o
Active object localization remains a critical challenge for robots, requiring efficient exploration of partially observable environments. However, state-of-the-art robot policies either struggle to generalize beyond demonstration datasets (e.g., imitation learning methods) or fail to generate physic…
Chain-of-thought (CoT) has emerged as a powerful technique to elicit reasoning in large language models and improve a variety of downstream tasks. CoT mainly demonstrates excellent performance in English, but its usage in low-resource languages is constrained due to poor language generalization. To…
In the domain of Document AI, parsing semi-structured image form is a crucial Key Information Extraction (KIE) task. The advent of pre-trained multimodal models significantly empowers Document AI frameworks to extract key information from form documents in different formats such as PDF, Word, and im…
Program synthesis aims to create accurate, executable programs from problem specifications, specifically from natural language descriptions in our context. Recent studies have leveraged the power of reinforcement learning (RL) in conjunction with large language models (LLMs), significantly enhancin…
Modular and composable transfer learning is an emerging direction in the field of Parameter Efficient Fine-Tuning, as it enables neural networks to better organize various aspects of knowledge, leading to improved cross-task generalization. In this paper, we introduce a novel approach Customized Pol…
The heavy ball momentum method is a commonly used technique for accelerating training processes in the machine learning community. However, empirical evidence suggests that the convergence of stochastic gradient descent (SGD) with heavy ball may slow down when the momentum hyperparameter approaches…
Parallel robots have passive joints and even passive limbs. Current calibration methods seldom consider the resulted identification and compensation properties and thus accuracy improvement is affected. This letter adopts finite and instantaneous screw (FIS) theory as mathematical tool. A generalize
Recently there has been a growing interest in category-level object pose and size estimation, and prevailing methods commonly rely on single view RGB-D images. However, one disadvantage of such methods is that they require accurate depth maps which cannot be produced by consumer-grade sensors. Furth…
Believable agents can empower interactive applications ranging from immersive environments to rehearsal spaces for interpersonal communication. Recently, generative agents have been proposed to simulate believable human behavior by using Large Language Models. However, the existing method heavily re…
STOchastic Recursive Momentum (STORM)-based algorithms have been widely developed to solve one to $K$-level ($K \geq 3$) stochastic optimization problems. Specifically, they use estimators to mitigate the biased gradient issue and achieve near-optimal convergence results. However, there is relativel…
Asynchronous stochastic gradient descent (ASGD) has evolved into an indispensable optimization algorithm for training modern large-scale distributed machine learning tasks. Therefore, it is imperative to explore the generalization performance of the ASGD algorithm. However, the existing results are…
Intermediate reasoning or acting steps have successfully improved large language models (LLMs) for handling various downstream natural language processing (NLP) tasks.When applying LLMs for code generation, recent works mainly focus on directing the models to articulate intermediate natural-language…
Embodying in vitro biological neural networks (BNNs) with robots to explore the rise of intelligence in these simpler models and to endow robots with biological intelligence has been attracting increasing attention in the fields of neuroscience and robotics. However, current research suffers from un
Tensegrity mobile robots are well-appraised for their high stiffness-to-mass ratio and superior structural compliance. However, traditional untethered tensegrity mobile robots usually have low velocity due to the large actuation force required by coupling effects among stiff struts and soft cables.
Due to the increasing computational demand of Deep Neural Networks (DNNs), companies and organizations have begun to outsource the training process. However, the externally trained DNNs can potentially be backdoor attacked. It is crucial to defend against such attacks, i.e, to postprocess a suspicio…
Enhancing model prediction confidence on target data is an important objective in Unsupervised Domain Adaptation (UDA). In this paper, we explore adversarial training on penultimate activations, i.e., input features of the final linear classification layer. We show that this strategy is more efficie…
Active Domain Adaptation (ADA) queries the labels of a small number of selected target samples to help adapting a model from a source domain to a target domain. The local context of queried data is important, especially when the domain gap is large. However, this has not been fully explored by exist…
Microfluidic printing provides a novel tool to facilitate the bulk assembly of cell-aligned microfibers for the fabrication of artificial skeletal muscle structure. However, due to the poor controllability for the deposition position of the microfiber, it is still difficult to realize the anisotropi
Sign Stochastic Gradient Descent (signSGD) is a communication-efficient stochastic algorithm that only uses the sign information of the stochastic gradient to update the model's weights. However, the existing convergence theory of signSGD either requires increasing batch sizes during training or ass…
Pose graph relaxation has become an indispensable addition to SLAM enabling efficient global registration of sensor reference frames under the objective of satisfying pair-wise relative transformation constraints. The latter may be given by incremental motion estimation or global place recognition.…
The generalization ability often determines the success of machine learning algorithms in practice. Therefore, it is of great theoretical and practical importance to understand and bound the generalization error of machine learning algorithms. In this paper, we provide the first generalization resul…
In this paper, we study the adaptive step size random walk gradient descent with momentum for decentralized optimization, in which the training samples are drawn dependently with each other. We establish theoretical convergence rates of the adaptive step size random walk gradient descent with moment…
Temporal difference (TD) learning with function approximations (linear functions or neural networks) has achieved remarkable empirical success, giving impetus to the development of finite-time analysis. As an accelerated version of TD, the adaptive TD has been proposed and proved to enjoy finite-tim…
A set of novel approaches for estimating epistemic uncertainty in deep neural networks with a single forward pass has recently emerged as a valid alternative to Bayesian Neural Networks. On the premise of informative representations, these deterministic uncertainty methods (DUMs) achieve strong perf…
"The waive of labels in the target domain makes Unsupervised Domain Adaptation (UDA) an attractive technique in many real-world applications, though it also brings great challenges as model adaptation becomes harder without labeled target data. In this paper, we address this issue by seeking compens…
Adapting to a continuously evolving environment is a safety-critical challenge inevitably faced by all autonomous-driving systems. Existing image- and video-based driving datasets, however, fall short of capturing the mutable nature of the real world. In this paper, we introduce the largest syntheti…
Unsupervised Domain Adaptation (UDA) aims to leverage a label-rich source domain to solve tasks on a related unlabeled target domain. It is a challenging problem especially when a large domain gap lies between the source and target domains. In this paper we propose a novel solution named SSRT (Safe…
We present an approach to learn voice-face representations from the talking face videos, without any identity labels. Previous works employ cross-modal instance discrimination tasks to establish the correlation of voice and face. These methods neglect the semantic content of different videos, introd…
In recent years, the Deep Learning Alternating Minimization (DLAM), which is actually the alternating minimization applied to the penalty form of the deep neutral networks training, has been developed as an alternative algorithm to overcome several drawbacks of Stochastic Gradient Descent (SGD) algo…
Many previous works have facilitated muscle cell (C2C12) alignment to form fiber-like cell structures. However, there still remains a challenge how to induce C2C12 myoblasts in the cell structures to differentiate into matured myocytes to form a functional muscle tissue, while external mechanical st…
Accelerating learning processes for complex tasks by leveraging previously learned tasks has been one of the most challenging problems in reinforcement learning, especially when the similarity between source and target tasks is low. This work proposes REPresentation And INstance Transfer (REPAINT) a…
The stability and generalization of stochastic gradient-based methods provide valuable insights into understanding the algorithmic performance of machine learning models. As the main workhorse for deep learning, the stochastic gradient descent has received a considerable amount of studies. Neverthel…
DeepRacer is a platform for end-to-end experimentation with RL and can be used to systematically investigate the key challenges in developing intelligent control systems. Using the platform, we demonstrate how a 1/18th scale car can learn to drive autonomously using RL with a monocular camera. It is…
Tissue engineering is trying to use modular tissue micro-rings to construct artificial biological microtubes as substitute of autologous tissue tubes to alleviate the shortage of donor sources. However, because of the lack of effective assembly strategies, it is still challenging to achieve high-spe…
In this work, we study the problem of multi-agent reinforcement learning (MARL) with model uncertainty, which is referred to as robust MARL. This is naturally motivated by some multi-agent applications where each agent may not have perfectly accurate knowledge of the model, e.g., all the reward func…
The research on rare cells makes a significant contribution to biology research and medical treatment for the application of diagnostic operation as well as prognoses treatment. Therefore, sorting them from heterogeneous mixtures is crucial and valuable. Traditional cell sorting methods featured wit…
The incremental aggregated gradient algorithm is popular in network optimization and machine learning research. However, the current convergence results require the objective function to be strongly convex. And the existing convergence rates are also limited to linear convergence. Due to the mathema…
In this paper, we consider a class of nonconvex problems with linear constraints appearing frequently in the area of image processing. We solve this problem by the penalty method and propose the iteratively reweighted alternating minimization algorithm. To speed up the algorithm, we also apply the c…
Deep learning is revolutionizing the mapping industry. Under lightweight human curation, computer has generated almost half of the roads in Thailand on Open- StreetMap (OSM) using high resolution aerial imagery. Bing maps are displaying 125 million computer generated building polygons in the U.S. Wh…
Microgrippers play a significant role in manipulation of micro-objects. To achieve dexterous and precise manipulation, a microgripper is required to be compactly designed and embedded with sensing feedback. Meanwhile, to convert the sensor position into displacement of the microgripper, the embedded…
This paper presents a new class of gradient methods for distributed machine learning that adaptively skip the gradient calculations to learn with reduced communication and computation. Simple rules are designed to detect slowly-varying gradients and, therefore, trigger the reuse of outdated grad…
Stochastic gradient methods are the workhorse (algorithms) of large-scale optimization problems in machine learning, signal processing, and other computational sciences and engineering. This paper studies Markov chain gradient descent, a variant of stochastic gradient descent where the random sample…
Asynchronous-parallel algorithms have the potential to vastly speed up algorithms by eliminating costly synchronization. However, our understanding of these algorithms is limited because the current convergence theory of asynchronous block coordinate descent algorithms is based on somewhat unrealist…
We investigate the problem of learning discrete graphical models in a differentially private way. Approaches to this problem range from privileged algorithms that conduct learning completely behind the privacy barrier to schemes that release private summary statistics paired with algorithms to learn…
In micromanipulation, lots of methods have been developed to manipulate objects in microscale. However, few of them can be applied in both the transportation and the rotation of the micro objects. In this paper, we present a novel method to realize the non-contact transportation and rotation of the…
Helical structure assembled by hydrogel microfibers is significant for culture of smooth muscle cells. However, the helical structure is only fabricated at the macroscale, while the fabrication of helical microstructure is still a challenge due to the lack of assembly method. In this paper, we propo…
In vitro tissues are significant biological substitute for drug test, cell morphogenesis exploration and organ transplantation. In this letter, a novel microrobotic bioassembly method is proposed to engineer 3-D cellular structure, which can be utilized to culture in vitro tissue with microstructura
Vascular-like microtube takes an important role in delivering oxygen and nutrient to keep cell alive in the generated tissue. In this paper, we present an automated micromanipulation system to assemble 2-D gel micro-rings to vascular-like microtubes by means of generating and controlling microbubble
Alginate hydrogel microfibers are a promising cell-laden module for three-dimensional (3-D) assembly to build cellular structures. However, it is still a challenge to manipulate them for microassembly. In this letter, we report a novel magnetic control method to handle this challenge. To enhance the
Fabrication of artificial blood vessels in micro scale significantly benefits the regeneration of functional human vascular networks. In this paper, we develop an efficient multi-microrobotic system with an innovative motorized sample holder (MSH) and two manipulators. Air is injected into the solut…
Collective graphical models (CGMs) are a formalism for inference and learning about a population of independent and identically distributed individuals when only noisy aggregate data are available. We highlight a close connection between approximate MAP inference in CGMs and marginal inference in st…
Due to the poor controllability in hydrogels, Hydrogels-based assembly to form larger 3D complex shapes is still a big challenge. In this paper, we have reported a novel “bottom-up” method to fabricate three-dimensional (3D) magnetic alginate microfibers (MAMs) assemblies with complex shapes. Specif…