75 accepted papers
Multi-Agent Path Finding (MAPF) in complex environments remains challenging due to high computational complexity, frequent conflicts, and realistic motion constraints. Most existing methods focus on discrete spaces or idealized omnidirectional models, often neglecting or partially considering nonhol
Although large-scale self-supervised learning (SSL) models like WavLM have achieved state-of-the-art performance in speech processing, their significant size impedes deployment on resource-constrained devices. While structured pruning is a key technique for model compression, existing methods typica…
In long-horizon tasks, recent agents based on Large Language Models (LLMs) face a significant challenge that sparse, outcome-based rewards make it difficult to assign credit to intermediate steps. Previous methods mainly focus on creating dense reward signals to guide learning, either through tradit…
High-Dynamic-Range Wide-Color-Gamut (HDR-WCG) technology is becoming increasingly widespread, driving a growing need for converting Standard Dynamic Range (SDR) content to HDR. Existing methods primarily rely on fixed tone mapping operators, which struggle to handle the diverse appearances and degra
Visual-language grounding aims to establish semantic correspondences between natural language and visual entities, enabling models to accurately identify and localize target objects based on textual instructions. Existing VLG approaches focus on coarse-grained, object-level localization, while tradi
Electroencephalography (EEG) plays a vital role in clinical and cognitive applications such as epilepsy diagnosis and emotion recognition. However, the low signal-to-noise ratio, inter-subject variability, and inherent non-stationarity of EEG signals present substantial modeling challenges. While re
Recent advances in generative AI have accelerated the production of ultra-high-resolution visual content. However, traditional image formats face significant limitations in efficient compression and real-time decoding, which restricts their applicability on end-user devices. Inspired by 3D Gaussian
The Mixture-of-Experts (MoE) structure scales the Transformer-based large language models (LLMs) and improves their performance with only the sub-linear increase in computation resources. Recently, a fine-grained DeepSeekMoE structure is proposed, which can further improve the computing efficiency o…
This paper introduces a fine-grained video tag classification algorithm that integrates knowledge graphs to identify specific TV series. A multimodal pretraining model extracts visual and textual features, feeding them into a multitask prediction model to classify type, genre, and entity tags. A sim…
Self-supervised learning (SSL) models for speaker verification (SV) have gained significant attention in recent years. However, existing SSL-based SV systems often struggle to capture local temporal dependencies and generalize across different tasks. In this paper, we propose context-aware multi-hea…
Audio deepfake detection (ADD) aims to verify the authenticity of audio. However, its performance declines sharply when facing significant domain discrepancies caused by unknown datasets. Unsupervised domain adaptation (UDA) has been applied to mitigate domain mismatch. However, as generative models…
Upcycled Mixture-of-Experts (MoE) models have shown great potential in various tasks by converting the original Feed-Forward Network (FFN) layers in pre-trained dense models into MoE layers. However, these models still suffer from significant parameter inefficiency due to the introduction of multipl…
The application scenarios of automated robots are undergoing a paradigm shift from structured environments to unstructured, complex settings. In highly constrained settings like factory inspections or disaster rescue, conventional steering systems show clear drawbacks. While the four-wheel independe
With the rapid advancement of intelligent manufacturing and the rise of emerging markets, global auto-mobile exports have surged, placing unprecedented demands on logistics infrastructure. Efficient coordination of multiple robots for vehicle autonomous transfer is essential in high-density storage
The Satellite Computing System (SCS) faces an increasing number of attacks. Although Moving Target Defense (MTD) can effectively mitigate attacks in ground networks, it is not well-suited for SCS due to the highly dynamic nature of both SCS traffic and attackers’ scanning behaviors. In this paper, w…
Understanding the internal mechanisms of transformer-based language models remains challenging. Mechanistic interpretability based on circuit discovery aims to reverse engineer neural networks by analyzing their internal processes at the level of computational subgraphs. In this paper, we revisit ex…
Automatic prompt engineering aims to enhance the generation quality of large language models (LLMs). Recent works utilize feedbacks generated from erroneous cases to guide the prompt optimization. During inference, they may further retrieve several semantically-related exemplars and concatenate them…
Self-reflection helps de-hallucinate Large Language Models (LLMs). However, the effectiveness of self-reflection remains insufficiently validated in the context of Small Language Models (SLMs), which exhibit limited semantic capacities. In particular, we demonstrate that the conventional self-reflec…
Multimodal Large Language Models (MLLMs) have shown impressive video content understanding capabilities but struggle with fine-grained motion comprehension. To comprehensively assess the motion understanding ability of existing MLLMs, we introduce FAVOR-Bench, which comprises 1,776 videos from both…
We present GEM, a Generalizable Ego-vision Multimodal world model that predicts future frames using a reference frame, sparse features, human poses, and ego-trajectories. Hence, our model has precise control over object dynamics, ego-agent motion and human poses. GEM generates paired RGB and depth o…
Regular temperature measurement of critical parts of an annealing furnace has always been a difficult task. Due to the harsh environment of high temperature, high noise, and darkness in the annealing furnace operation area, unmanned vehicles equipped with the RGB-T semantic segmentation model are us
Previous fake speech datasets were constructed from a defender’s perspective to develop countermeasure (CM) systems without considering diverse motivations of attackers. To better align with real-life scenarios, we created LlamaPartialSpoof, a 130-hour dataset that contains both fully and partially…
Emotion Cause Triplet Extraction in Multimodal Conversations (MECTEC) has recently gained significant attention in social media analysis, aiming to extract emotion utterances, cause utterances, and emotion categories simultaneously. However, the scarcity of related datasets, with only one published…
Transformer-based language models have achieved significant success; however, their internal mechanisms remain largely opaque due to the complexity of non-linear interactions and high-dimensional operations. While previous studies have demonstrated that these models implicitly embed reasoning trees,…
Diffusion models have demonstrated impressive performance in face restoration. Yet, their multi-step inference process remains computationally intensive, limiting their applicability in real-world scenarios. Moreover, existing methods often struggle to generate face images that are harmonious, reali…
While Large Language Models (LLMs) demonstrate strong performance across domains, their long-context capabilities are limited by transient neural activations causing information decay and unstructured feed-forward network (FFN) weights leading to semantic fragmentation. Inspired by the brain’s worki…
Although Coordinate-MLP-based implicit neural representations have excelled in representing radiance fields, 3D shapes, and images, their application to audio signals remains underexplored. To fill this gap, we investigate existing implicit neural representations, from which we extract 3 types of po…
We propose SC-Captioner, a reinforcement learning framework that enables the self-correcting capability of image caption models. Our crucial technique lies in the design of the reward function to incentivize accurate caption corrections. Specifically, the predicted and reference captions are decompo…
Large Language Models have revolutionized numerous tasks with their remarkable efficacy. However, editing these models, crucial for rectifying outdated or erroneous information, often leads to a complex issue known as the ripple effect in the hidden space. While difficult to detect, this effect can…
Audio-visual navigation has received considerable attention in recent years. However, the majority of related investigations have focused on single sound-source scenarios. Studies in this field for multiple sound-source scenarios remain underexplored due to the limitations of two aspects. First, the…
Transformer-based architectures have been proven successful in detecting 3D objects from point clouds. However, the quadratic complexity of the attention mechanism struggles to encode rich information as point cloud resolution increases. Recently, state space models (SSM) such as Mamba have gained g…
"Current visual generation methods can produce high-quality videos guided by text prompts. However, effectively controlling object dynamics remains a challenge. This work explores audio as a cue to generate temporally synchronized image animations. We introduce Audio-Synchronized Visual Animation ()…
Conventional copy-paste augmentations generate new training instances by concatenating existing utterances to increase the amount of data for neural network training. However, the direct application of copy-paste augmentation for anti-spoofing is problematic. This paper refines the copy-paste augmen…
Detecting evidence within the context is a key step in the process of reasoning task. Evaluating and enhancing the capabilities of LLMs in evidence detection will strengthen context-based reasoning performance. This paper proposes a benchmark called DetectBench for verifying the ability to detect an…
In the labor-intensive Roll-On/Roll-Off (RO/RO) terminal environment, research on vehicle transport robots with mobility, stability, and reliability is receiving increasing attention. This paper presents a novel control framework for a Straddle-Type Dual-Body vehicle transfer robot. Initially, fine…
New Natural Langauge Process~(NLP) benchmarks are urgently needed to align with the rapid development of large language models (LLMs). We present Xiezhi, the most comprehensive evaluation suite designed to assess holistic domain knowledge.Xiezhi comprises multiple-choice questions across 516 diverse…
Event skeleton generation, aiming to induce an event schema skeleton graph with abstracted event nodes and their temporal relations from a set of event instance graphs, is a critical step in the temporal complex event schema induction task. Existing methods effectively address this task from a graph…
Subgraph matching is a fundamental building block for graph-based applications and is challenging due to its high-order combinatorial nature. Existing studies usually tackle it by combinatorial optimization or learning-based methods. However, they suffer from exponential computational costs or searc…
Neural architecture search (NAS) for Graph neural networks (GNNs), called NAS-GNNs, has achieved significant performance over manually designed GNN architectures. However, these methods inherit issues from the conventional NAS methods, such as high computational cost and optimization difficulty. Mor…
A Cable-Driven Parallel Robot (CDPR) with Mobile Bases (MBs) can modify its geometric architecture and is suitable for manipulation tasks in constrained environments. In manipulation tasks, a CDPR with MBs inevitably encounters obstacles, including dynamic obstacles. However, the high dimensional st
Second-order optimization algorithms exhibit excellent convergence properties for training deep learning models, but often incur significant computation and memory overheads. This can result in lower training efficiency than the first-order counterparts such as stochastic gradient descent (SGD). In…
Graph neural networks aim to learn representations for graph-structured data and show impressive performance in node classification. Recently, many methods have studied the representations of GNNs from the perspective of optimization goals and spectral graph theory. However, the feature space that d…
It is widely agreed that reference-based super-resolution (RefSR) achieves superior results by referring to similar high quality images, compared to single image super-resolution (SISR). Intuitively, the more references, the better performance. However, previous RefSR methods have all focused on sin…
Multimodal semantic understanding often has to deal with uncertainty, which means the obtained messages tend to refer to multiple targets. Such uncertainty is problematic for our interpretation, including inter- and intra-modal uncertainty. Little effort has studied the modeling of this uncertainty,…
Recent advances in pre-trained language models (PLMs) have facilitated the development ofcommonsense reasoning tasks. However, existing methods rely on multi-hop knowledgeretrieval and thus suffer low accuracy due toembedded noise in the acquired knowledge. In addition, these methods often attain hi…
Optimal transport (OT) can be applied to cross-domain adaptation in speaker verification (SV) by converting speakers' probability distributions from source to target domains. However, in scenarios involving over-massive categories (speakers) or difficult samples in discrimination, OT often has diffi…
Deep learning methods have demonstrated promising performance on the NP-hard Graph Matching (GM) problems. However, the state-of-the-art methods usually require the ground-truth labels, which may take extensive human efforts or be impractical to collect. In this paper, we present a robust self-super…
Large-scale pre-trained language models (PLMs) bring new opportunities to challenging problems, especially those that need high-level intelligence, such as the math word problem (MWPs). However, directly applying existing PLMs to MWPs can fail as the generation process lacks sufficient supervision a…
Among various cross-modal retrieval methods, the supervised methods achieve the best performance by exploiting the semantic labels. However, in realistic applications, the data are not always complete with labels and full multi-modal data, which makes these methods hard to be used. In this paper, we…
This paper presents CCRobot-V, the fifth version of CCRobot, a cooperative serial multi-robot system for bridge cable inspection and maintenance that uses silkworm-like locomotion to climb the entire length of super-long stay cable at high speeds while carrying heavy inspection/maintenance equipment…
Automatic speaker verification (ASV) systems, which determine whether two speeches are from the same speaker, mainly focus on verification accuracy while ignoring inference speed. However, in real applications, both inference speed and verification accuracy are essential. This study proposes cross-s…
Recent years have witnessed a growing interest in online RGB-D 3D reconstruction. On the premise of ensuring the reconstruction accuracy with noisy depth scans, making the system scalable to various environments is still challenging. In this paper, we devote our efforts to try to fill in this resear…
For exact 3D shape correspondence (matching or alignment), i.e., the task of matching each point on a shape to its exact corresponding point on the other shape (or to be more specific, matching at geodesic error 0), most existing methods do not perform well due to two main problems. First, on nearly…
Federated Learning (FL) is an emerging distributed learning paradigm under privacy constraint. Data heterogeneity is one of the main challenges in FL, which results in slow convergence and degraded performance. Most existing approaches only tackle the heterogeneity challenge by restricting the local…
Multi-robot coarse-to-fine exploration in unknown environments makes great sense in many application fields like search and rescue. For different stages of the task, robots need to extract information from the environment discriminately, which can improve their decision-making capability. To this en
Semi-supervised video action recognition tends to enable deep neural networks to achieve remarkable performance even with very limited labeled data. However, existing methods are mainly transferred from current image-based methods (e.g., FixMatch). Without specifically utilizing the temporal dynamic…
Achieving good performance on few-shot or zero-shot datasets has been a long-term challenge for NER. The conventional semantic transfer approaches on NER will decrease model performance when the semantic distribution is quite different, especially in Chinese few-shot NER. Recently, prompt-tuning has…
"Reference-based image super-resolution (RefSR) is a promising SR branch and has shown great potential in overcoming the limitations of single image super-resolution. While previous state-of-the-art RefSR methods mainly focus on improving the efficacy and robustness of reference feature transfer, it…
To overcome the overfitting issue of dehazing models trained on synthetic hazy-clean image pairs, many recent methods attempted to improve models' generalization ability by training on unpaired data. Most of them simply formulate dehazing and rehazing cycles, yet ignore the physical properties of th…
The realistic simulation of underwater scenes has important significance for many researches related to underwater vision, such as underwater image restoration, underwater moving object monitoring, etc. To date, however, the existing underwater scene simulation pipelines are either too complicated d…
We propose a new paradigm for zero-shot learners that is format agnostic, i.e., it is compatible with any format and applicable to a list of language tasks, such as text classification, commonsense reasoning, coreference resolution, and sentiment analysis. Zero-shot learning aims to train a model on…
Federated Learning (FL) aims to establish a shared model across decentralized clients under the privacy-preserving constraint. Despite certain success, it is still challenging for FL to deal with non-IID (non-independent and identical distribution) client data, which is a general scenario in real-wo…
Retrieving occlusion relation among objects in a single image is challenging due to sparsity of boundaries in image. We observe two key issues in existing works: firstly, lack of an architecture which can exploit the limited amount of coupling in the decoder stage between the two subtasks, namely oc…
Missing data is a common problem in multimodal and multi-view learning. It raises a critical challenge for most multimodal algorithms, which are unable to deal with incomplete datasets. Rather than discarding entries with missing modalities, this paper aims to reconstruct the complete image-based mu…
Thanks to their high carrying capacity and strong maneuverability, carlike robots which move with non-holonomic constraints, are frequently utilized in numerous coverage operation fields. In such fields, the robots need to complete the coverage task via autonomous path planning and tracking, which i
Master-slave control is a common form of human-robot interaction for robotic surgery. To ensure seamless and intuitive control, a mechanism of self-adaptive motion scaling during teleoperaton is proposed in this letter. The operator can retain precise control when conducting delicate or complex mani
Snake-like robots have numerous applications in minimally invasive surgery. One important research topic of snake-like robots is the flexible joint mechanism and its actuation. This letter describes the design and fabrication of a new type of flexible joint mechanism that is enabled by metal powder
Master manipulators represent a key component of Robot-Assisted Minimally Invasive Surgery (RAMIS). In this paper, an Analytic Hierarchy Process (AHP) method is used to construct an effective workspace analysis framework, which can assist the configuration selection and design evaluation of a portab…
Suturing and knot tying in a confined space is a technically challenging yet clinically demanding task in minimally invasive surgery, which requires the use of highly articulated instruments passing through small incisions on the patient's body. Manually operating such instruments is usually very di
Task autonomy is an important consideration for the development of future surgical robots. For robot-assisted anastomosis, suture thread detection is a prerequisite for subsequent robot manipulation. Previous works on automatic thread detection are focused on the learning of the models with specific…
Estimating the three-dimensional (3D) position of microrobots is necessary in order to develop closed-loop control techniques and to improve the user's 3D perception in the micro-scale. This paper describes a depth estimation method based on supervised learning for optically transparent microrobots…
This paper presents a versatile robotic system for sewing a 3D structured object. Leveraging on using a customized robotic sewing device and closed-loop visual servoing control, an all-in-one solution for sewing personalized stent graft is demonstrated. Stitch size planning and automated knot tying…
Robot-assisted minimally invasive surgery can benefit from the automation of common, repetitive or well-defined but ergonomically difficult tasks. One such task is the scanning of a pick-up endomicroscopy probe over a complex, undulating tissue surface to enhance the effective field-of-view through…
Six degree-of-freedom (DoF) pose feedback is essential for the development of closed-loop control techniques for microrobotics. This paper presents two methods for depth estimation of transparent microrobots inside an Optical Tweezers (OT) setup using image sharpness measurements and model-based tra…
Energy-efficient signal processing systems require estimation methods operating on data collected with low-complexity devices. Using analog-to-digital converters (ADC) with 1-bit amplitude resolution has been identified as a possible option in order to obtain low power consumption. The 1-bit perform…