Teng Li

Yitong Sun, Yao Huang, Teng Li et al.

Mixture-of-Experts (MoE) architectures scale Large Language Models (LLMs) efficiently, enabling greater capacity with reduced computational cost by dynamically routing inputs to relevant experts, yet introduce a critical vulnerability: Safety Sparsity, where safety capabilities concentrate in few experts, making them susceptible to adversarial bypassing. Meanwhile, conventional alignment methods uniformly adapt all parameters, ignoring their functional differences and inadvertently degrading performances. To address these challenges, we propose MESA (MoE Safety Alignment), a targeted alignment framework for MoE-based LLMs that strategically decentralizes safety responsibility to maximize coverage while minimizing interference with utility. Based on Optimal Transport (OT) theory, MESA operates through two mechanisms: (1) Expert Capacity Reallocation uses a transport cost matrix to distribute safety duties to the most cost-effective experts, and (2) Dynamic Routing Refinement constrains the router to precisely activate these decentralized modules. Experiments show that MESA achieves robust defensive performance against varied harmful benchmarks while preserving helpfulness. Code is available at https://github.com/lorraine021/MESA.

Xiaoyang Huang, Yi Zhang, Bingbing Ni et al.

Recent years we have witnessed rapid development in NeRF-based image rendering due to its high quality. However, point clouds rendering is somehow less explored. Compared to NeRF-based rendering which suffers from dense spatial sampling, point clouds rendering is naturally less computation intensive, which enables its deployment in mobile computing device. In this work, we focus on boosting the image quality of point clouds rendering with a compact model design. We first analyze the adaption of the volume rendering formulation on point clouds. Based on the analysis, we simplify the NeRF representation to a spatial mapping function which only requires single evaluation per pixel. Further, motivated by ray marching, we rectify the the noisy raw point clouds to the estimated intersection between rays and surfaces as queried coordinates, which could avoid \textit{spatial frequency collapse} and neighbor point disturbance. Composed of rasterization, spatial mapping and the refinement stages, our method achieves the state-of-the-art performance on point clouds rendering, outperforming prior works by notable margins, with a smaller model size. We obtain a PSNR of 31.74 on NeRF-Synthetic, 25.88 on ScanNet and 30.81 on DTU. Code and data are publicly available at https://github.com/seanywang0408/RadianceMapping.

Xiaoyang Huang, Jiancheng Yang, Yanjun Wang et al.

3D shape analysis has been widely explored in the era of deep learning. Numerous models have been developed for various 3D data representation formats, e.g., MeshCNN for meshes, PointNet for point clouds and VoxNet for voxels. In this study, we present Representation-Agnostic Shape Fields (RASF), a generalizable and computation-efficient shape embedding module for 3D deep learning. RASF is implemented with a learnable 3D grid with multiple channels to store local geometry. Based on RASF, shape embeddings for various 3D shape representations (point clouds, meshes and voxels) are retrieved by coordinate indexing. While there are multiple ways to optimize the learnable parameters of RASF, we provide two effective schemes among all in this paper for RASF pre-training: shape reconstruction and normal estimation. Once trained, RASF becomes a plug-and-play performance booster with negligible cost. Extensive experiments on diverse 3D representation formats, networks and applications, validate the universal effectiveness of the proposed RASF. Code and pre-trained models are publicly available https://github.com/seanywang0408/RASF

Xiaoyang Huang, Yanjun Wang, Yang Liu et al.

Spatial audio, which focuses on immersive 3D sound rendering, is widely applied in the acoustic industry. One of the key problems of current spatial audio rendering methods is the lack of personalization based on different anatomies of individuals, which is essential to produce accurate sound source positions. In this work, we address this problem from an interdisciplinary perspective. The rendering of spatial audio is strongly correlated with the 3D shape of human bodies, particularly ears. To this end, we propose to achieve personalized spatial audio by reconstructing 3D human ears with single-view images. First, to benchmark the ear reconstruction task, we introduce AudioEar3D, a high-quality 3D ear dataset consisting of 112 point cloud ear scans with RGB images. To self-supervisedly train a reconstruction model, we further collect a 2D ear dataset composed of 2,000 images, each one with manual annotation of occlusion and 55 landmarks, named AudioEar2D. To our knowledge, both datasets have the largest scale and best quality of their kinds for public use. Further, we propose AudioEarM, a reconstruction method guided by a depth estimation network that is trained on synthetic data, with two loss functions tailored for ear data. Lastly, to fill the gap between the vision and acoustics community, we develop a pipeline to integrate the reconstructed ear mesh with an off-the-shelf 3D human body and simulate a personalized Head-Related Transfer Function (HRTF), which is the core of spatial audio rendering. Code and data are publicly available at https://github.com/seanywang0408/AudioEar.

Yi Zhang, Xiaoyang Huang, Bingbing Ni et al.

We develop an effective point cloud rendering pipeline for novel view synthesis, which enables high fidelity local detail reconstruction, real-time rendering and user-friendly editing. In the heart of our pipeline is an adaptive frequency modulation module called Adaptive Frequency Net (AFNet), which utilizes a hypernetwork to learn the local texture frequency encoding that is consecutively injected into adaptive frequency activation layers to modulate the implicit radiance signal. This mechanism improves the frequency expressive ability of the network with richer frequency basis support, only at a small computational budget. To further boost performance, a preprocessing module is also proposed for point cloud geometry optimization via point opacity estimation. In contrast to implicit rendering, our pipeline supports high-fidelity interactive editing based on point cloud manipulation. Extensive experimental results on NeRF-Synthetic, ScanNet, DTU and Tanks and Temples datasets demonstrate the superior performances achieved by our method in terms of PSNR, SSIM and LPIPS, in comparison to the state-of-the-art.

Zefan Li, Bingbing Ni, Teng Li et al.

The great success neural networks have achieved is inseparable from the application of gradient-descent (GD) algorithms. Based on GD, many variant algorithms have emerged to improve the GD optimization process. The gradient for back-propagation is apparently the most crucial aspect for the training of a neural network. The quality of the calculated gradient can be affected by multiple aspects, e.g., noisy data, calculation error, algorithm limitation, and so on. To reveal gradient information beyond gradient descent, we introduce a framework (\textbf{GCGD}) to perform gradient correction. GCGD consists of two plug-in modules: 1) inspired by the idea of gradient prediction, we propose a \textbf{GC-W} module for weight gradient correction; 2) based on Neural ODE, we propose a \textbf{GC-ODE} module for hidden states gradient correction. Experiment results show that our gradient correction framework can effectively improve the gradient quality to reduce training epochs by $\sim$ 20\% and also improve the network performance.

Yuanhuiyi Lyu, Kaiyu Lei, Ziqiao Weng et al.

Reasoning-based text-to-image (T2I) generation requires models to interpret complex prompts accurately. Existing reasoning frameworks can be broadly categorized into two types: (1) Text-Only Reasoning, which is computationally efficient but lacks access to visual context, often resulting in the omission of critical spatial and visual elements; and (2) Text-Image Interleaved Reasoning, which leverages a T2I generator to provide visual references during the reasoning process. While this approach enhances visual grounding, it incurs substantial computational costs and constrains the reasoning capacity of MLLMs to the representational limitations of the generator. To this end, we propose StruVis, a novel framework that enhances T2I generation through Thinking with Structured Vision. Instead of relying on intermediate image generation, StruVis employs text-based structured visual representations as intermediate reasoning states, thereby enabling the MLLM to effectively "perceive" visual structure within a purely text-based reasoning process. Powered by this, the reasoning potential for T2I generation of the MLLM is unlocked through structured-vision-guided reasoning. Additionally, as a generator-agnostic reasoning framework, our proposed StruVis can be seamlessly integrated with diverse T2I generators and efficiently enhance their performance in reasoning-based T2I generation. Extensive experiments demonstrate that StruVis achieves significant performance improvements on reasoning-based T2I benchmarks, e.g., a 4.61% gain on T2I-ReasonBench and a 4% gain on WISE.

Bowen Lu, Liangqiang Yang, Teng Li

Graph Neural Networks (GNNs) have achieved impressive performance in graph-related tasks. However, they suffer from poor generalization on out-of-distribution (OOD) data, as they tend to learn spurious correlations. Such correlations present a phenomenon that GNNs fail to stably learn the mutual information between prediction representations and ground-truth labels under OOD settings. To address these challenges, we formulate a causal graph starting from the essence of node classification, adopt backdoor adjustment to block non-causal paths, and theoretically derive a lower bound for improving OOD generalization of GNNs. To materialize these insights, we further propose a novel approach integrating causal representation learning and a loss replacement strategy. The former captures node-level causal invariance and reconstructs graph posterior distribution. The latter introduces asymptotic losses of the same order to replace the original losses. Extensive experiments demonstrate the superiority of our method in OOD generalization and effectively alleviating the phenomenon of unstable mutual information learning.

Hao Zhang, Zhendong Pang, Jiangpeng Wang et al.

Deep neural networks (DNNs) that tackle the time series classification (TSC) task have provided a promising framework in signal processing. In real-world applications, as a data-driven model, DNNs are suffered from insufficient data. Few-shot learning has been studied to deal with this limitation. In this paper, we propose a novel few-shot learning framework through data augmentation, which involves transformation through the time-frequency domain and the generation of synthetic images through random erasing. Additionally, we develop a sequence-spectrogram neural network (SSNN). This neural network model composes of two sub-networks: one utilizing 1D residual blocks to extract features from the input sequence while the other one employing 2D residual blocks to extract features from the spectrogram representation. In the experiments, comparison studies of different existing DNN models with/without data augmentation are conducted on an amyotrophic lateral sclerosis (ALS) dataset and a wind turbine fault (WTF) dataset. The experimental results manifest that our proposed method achieves 93.75% F1 score and 93.33% accuracy on the ALS datasets while 95.48% F1 score and 95.59% accuracy on the WTF datasets. Our methodology demonstrates its applicability of addressing the few-shot problems for time series classification.

Guangcong Zheng, Teng Li, Xianpan Zhou et al.

Recent advances in camera-controllable video generation have been constrained by the reliance on static-scene datasets with relative-scale camera annotations, such as RealEstate10K. While these datasets enable basic viewpoint control, they fail to capture dynamic scene interactions and lack metric-scale geometric consistency-critical for synthesizing realistic object motions and precise camera trajectories in complex environments. To bridge this gap, we introduce the first fully open-source, high-resolution dynamic-scene dataset with metric-scale camera annotations in https://github.com/ZGCTroy/RealCam-Vid.

Zening Lin, Jiapeng Wang, Teng Li et al.

Document pair extraction aims to identify key and value entities as well as their relationships from visually-rich documents. Most existing methods divide it into two separate tasks: semantic entity recognition (SER) and relation extraction (RE). However, simply concatenating SER and RE serially can lead to severe error propagation, and it fails to handle cases like multi-line entities in real scenarios. To address these issues, this paper introduces a novel framework, PEneo (Pair Extraction new decoder option), which performs document pair extraction in a unified pipeline, incorporating three concurrent sub-tasks: line extraction, line grouping, and entity linking. This approach alleviates the error accumulation problem and can handle the case of multi-line entities. Furthermore, to better evaluate the model's performance and to facilitate future research on pair extraction, we introduce RFUND, a re-annotated version of the commonly used FUNSD and XFUND datasets, to make them more accurate and cover realistic situations. Experiments on various benchmarks demonstrate PEneo's superiority over previous pipelines, boosting the performance by a large margin (e.g., 19.89%-22.91% F1 score on RFUND-EN) when combined with various backbones like LiLT and LayoutLMv3, showing its effectiveness and generality. Codes and the new annotations are available at https://github.com/ZeningLin/PEneo.

Xinzge Gao, Chuanrui Hu, Bin Chen et al.

Multimodal large language models (MLLMs) are attracting growing attention in the development of Graphical User Interface (GUI) agents. Existing approaches often rely on historical screenshots or actions to implicitly represent the task state. This reliance poses challenges for GUI agents in accurately understanding task states and underscores the absence of effective mechanisms to store critical information in complex and lengthy cross-app tasks. To address these challenges, we propose Chain-of-Memory (CoM), a novel approach for explicitly modeling short-term and long-term memory in GUI agents. CoM achieves this by capturing action descriptions, integrating task-relevant screen information, and maintaining a dedicated memory module to store and manage this information. By leveraging explicit memory representations, CoM enables GUI agents to better understand task states and retain critical historical information persistently. To equip GUI agents with memory management capabilities and evaluate the effectiveness of CoM, we developed the GUI Odyssey-CoM, a dataset comprising 111k screen-action pairs annotated with Chain-of-Memory. Experimental results demonstrate that CoM significantly improves GUI agents' performance in cross-application tasks. Additionally, GUI Odyssey-CoM enables 7B models to achieve memory management capabilities comparable to 72B models. The dataset and code will be open-sourced.

Linnan Wang, Saining Xie, Teng Li et al.

Neural Architecture Search (NAS) has emerged as a promising technique for automatic neural network design. However, existing MCTS based NAS approaches often utilize manually designed action space, which is not directly related to the performance metric to be optimized (e.g., accuracy), leading to sample-inefficient explorations of architectures. To improve the sample efficiency, this paper proposes Latent Action Neural Architecture Search (LaNAS), which learns actions to recursively partition the search space into good or bad regions that contain networks with similar performance metrics. During the search phase, as different action sequences lead to regions with different performance, the search efficiency can be significantly improved by biasing towards the good regions. On three NAS tasks, empirical results demonstrate that LaNAS is at least an order more sample efficient than baseline methods including evolutionary algorithms, Bayesian optimizations, and random search. When applied in practice, both one-shot and regular LaNAS consistently outperform existing results. Particularly, LaNAS achieves 99.0% accuracy on CIFAR-10 and 80.8% top1 accuracy at 600 MFLOPS on ImageNet in only 800 samples, significantly outperforming AmoebaNet with 33x fewer samples. Our code is publicly available at https://github.com/facebookresearch/LaMCTS.

Guangcong Zheng, Teng Li, Rui Jiang et al.

Recent advancements have integrated camera pose as a user-friendly and physics-informed condition in video diffusion models, enabling precise camera control. In this paper, we identify one of the key challenges as effectively modeling noisy cross-frame interactions to enhance geometry consistency and camera controllability. We innovatively associate the quality of a condition with its ability to reduce uncertainty and interpret noisy cross-frame features as a form of noisy condition. Recognizing that noisy conditions provide deterministic information while also introducing randomness and potential misguidance due to added noise, we propose applying epipolar attention to only aggregate features along corresponding epipolar lines, thereby accessing an optimal amount of noisy conditions. Additionally, we address scenarios where epipolar lines disappear, commonly caused by rapid camera movements, dynamic objects, or occlusions, ensuring robust performance in diverse environments. Furthermore, we develop a more robust and reproducible evaluation pipeline to address the inaccuracies and instabilities of existing camera control metrics. Our method achieves a 25.64% improvement in camera controllability on the RealEstate10K dataset without compromising dynamics or generation quality and demonstrates strong generalization to out-of-domain images. Training and inference require only 24GB and 12GB of memory, respectively, for 16-frame sequences at 256x256 resolution. We will release all checkpoints, along with training and evaluation code. Dynamic videos are best viewed at https://zgctroy.github.io/CamI2V.

Yangfu Li, Yuning Gong, Hongjian Zhan et al.

Despite the success of Large-Vision Language Models (LVLMs), general optimization objectives (e.g., standard MLE) fail to constrain visual trajectories, leading to language bias and hallucination. To mitigate this, current methods introduce geometric priors from visual experts as additional supervision. However, we observe that such supervision is typically suboptimal: it is biased toward geometric precision and offers limited reasoning utility. To bridge this gap, we propose Perceptual Flow Network (PFlowNet), which eschews rigid alignment with the expert priors and achieves interpretable yet more effective visual reasoning. Specifically, PFlowNet decouples perception from reasoning to establish a self-conditioned generation process. Based on this, it integrates multi-dimensional rewards with vicinal geometric shaping via variational reinforcement learning, thereby facilitating reasoning-oriented perceptual behaviors while preserving visual reliability. PFlowNet delivers a provable performance guarantee and competitive empirical results, particularly setting new SOTA records on V* Bench (90.6%) and MME-RealWorld-lite (67.0%).

Wenqiang Sun, Teng Li, Zehong Lin et al.

Recently, text-to-image diffusion models have demonstrated impressive ability to generate high-quality images conditioned on the textual input. However, these models struggle to accurately adhere to textual instructions regarding spatial layout information. While previous research has primarily focused on aligning cross-attention maps with layout conditions, they overlook the impact of the initialization noise on the layout guidance. To achieve better layout control, we propose leveraging a spatial-aware initialization noise during the denoising process. Specifically, we find that the inverted reference image with finite inversion steps contains valuable spatial awareness regarding the object's position, resulting in similar layouts in the generated images. Based on this observation, we develop an open-vocabulary framework to customize a spatial-aware initialization noise for each layout condition. Without modifying other modules except the initialization noise, our approach can be seamlessly integrated as a plug-and-play module within other training-free layout guidance frameworks. We evaluate our approach quantitatively and qualitatively on the available Stable Diffusion model and COCO dataset. Equipped with the spatial-aware latent initialization, our method significantly improves the effectiveness of layout guidance while preserving high-quality content.

Teng Li, Guangcong Zheng, Rui Jiang et al.

Recent advancements in camera-trajectory-guided image-to-video generation offer higher precision and better support for complex camera control compared to text-based approaches. However, they also introduce significant usability challenges, as users often struggle to provide precise camera parameters when working with arbitrary real-world images without knowledge of their depth nor scene scale. To address these real-world application issues, we propose RealCam-I2V, a novel diffusion-based video generation framework that integrates monocular metric depth estimation to establish 3D scene reconstruction in a preprocessing step. During training, the reconstructed 3D scene enables scaling camera parameters from relative to metric scales, ensuring compatibility and scale consistency across diverse real-world images. In inference, RealCam-I2V offers an intuitive interface where users can precisely draw camera trajectories by dragging within the 3D scene. To further enhance precise camera control and scene consistency, we propose scene-constrained noise shaping, which shapes high-level noise and also allows the framework to maintain dynamic and coherent video generation in lower noise stages. RealCam-I2V achieves significant improvements in controllability and video quality on the RealEstate10K and out-of-domain images. We further enables applications like camera-controlled looping video generation and generative frame interpolation. Project page: https://zgctroy.github.io/RealCam-I2V.

Teng Li, Stephen Wu, Yong Huang et al.

The health condition of components in civil infrastructures can be described by various discrete states according to their performance degradation. Inferring these states from measurable responses is typically an ill-posed inverse problem. Although Bayesian methods are well-suited to tackle such problems, computing the posterior probability density function (PDF) presents challenges. The likelihood function cannot be analytically formulated due to the unclear relationship between discrete states and structural responses, and the high-dimensional state parameters resulting from numerous components severely complicates the computation of the marginal likelihood function. To address these challenges, this study proposes a novel Bayesian inversion paradigm for discrete variables based on Probabilistic Graphical Models (PGMs). The Markov networks are employed as modeling tools, with model parameters learned from data and structural topology prior. It has been proved that inferring this PGM produces the same probabilistic estimation as the posterior PDF derived from Bayesian inference, which effectively solves the above challenges. The inference is accomplished by Graph Neural Networks (GNNs), and a graph property-based GNN training strategy is developed to enable accurate inference across varying graph scales, thereby significantly reducing the computational overhead in high-dimensional problems. Both synthetic and experimental data are used to validate the proposed framework

Teng Li, Quanfeng Lu, Lirui Zhao et al.

Unified image understanding and generation has emerged as a promising paradigm in multimodal artificial intelligence. Despite recent progress, the optimal architectural design for such unified models remains an open challenge. In this work, we start by analyzing the modality alignment behaviors of task-specific expert models for understanding and generation, as well as current unified models. Our analysis reveals a crucial observation: understanding tasks benefit from a progressively increasing modality alignment across network depth, which helps build up semantic information for better comprehension; In contrast, generation tasks follow a different trend: modality alignment increases in the early layers but decreases in the deep layers to recover spatial details. These divergent alignment patterns create a fundamental conflict in fully shared Transformer backbones, where a uniform representational flow often leads to performance compromises across two tasks. Motivated by this finding, we introduce UniFork, a novel Y-shaped architecture that shares the shallow layers for cross-task representation learning, while employing task-specific branches in deeper layers to avoid task interference. This design effectively balances shared learning and task specialization. Through extensive ablation experiments, we demonstrate that Unifork consistently outperforms conventional fully shared Transformer architectures, and achieves performance on par with or better than task-specific models.

Teng Li, Xingjun Ma, Yu-Gang Jiang

Transferable adversarial examples highlight the vulnerability of deep neural networks (DNNs) to imperceptible perturbations across various real-world applications. While there have been notable advancements in untargeted transferable attacks, targeted transferable attacks remain a significant challenge. In this work, we focus on generative approaches for targeted transferable attacks. Current generative attacks focus on reducing overfitting to surrogate models and the source data domain, but they often overlook the importance of enhancing transferability through additional semantics. To address this issue, we introduce a novel plug-and-play module into the general generator architecture to enhance adversarial transferability. Specifically, we propose a \emph{Semantic Injection Module} (SIM) that utilizes the semantics contained in an additional guiding image to improve transferability. The guiding image provides a simple yet effective method to incorporate target semantics from the target class to create targeted and highly transferable attacks. Additionally, we propose new loss formulations that can integrate the semantic injection module more effectively for both targeted and untargeted attacks. We conduct comprehensive experiments under both targeted and untargeted attack settings to demonstrate the efficacy of our proposed approach.

Ao Liu, Jing Chen, Ruiying Du et al.

The rapid expansion of Internet of Things (IoT) has resulted in vast, heterogeneous graphs that capture complex interactions among devices, sensors, and systems. Efficient analysis of these graphs is critical for deriving insights in IoT scenarios such as smart cities, industrial IoT, and intelligent transportation systems. However, the scale and diversity of IoT-generated data present significant challenges, and existing methods often struggle with preserving the structural integrity and semantic richness of these complex graphs. Many current approaches fail to maintain the balance between computational efficiency and the quality of the insights generated, leading to potential loss of critical information necessary for accurate decision-making in IoT applications. We introduce HeteroSample, a novel sampling method designed to address these challenges by preserving the structural integrity, node and edge type distributions, and semantic patterns of IoT-related graphs. HeteroSample works by incorporating the novel top-leader selection, balanced neighborhood expansion, and meta-path guided sampling strategies. The key idea is to leverage the inherent heterogeneous structure and semantic relationships encoded by meta-paths to guide the sampling process. This approach ensures that the resulting subgraphs are representative of the original data while significantly reducing computational overhead. Extensive experiments demonstrate that HeteroSample outperforms state-of-the-art methods, achieving up to 15% higher F1 scores in tasks such as link prediction and node classification, while reducing runtime by 20%.These advantages make HeteroSample a transformative tool for scalable and accurate IoT applications, enabling more effective and efficient analysis of complex IoT systems, ultimately driving advancements in smart cities, industrial IoT, and beyond.

Teng Li, Ziyuan Huang, Cong Chen et al.

We propose TC-AE, a ViT-based architecture for deep compression autoencoders. Existing methods commonly increase the channel number of latent representations to maintain reconstruction quality under high compression ratios. However, this strategy often leads to latent representation collapse, which degrades generative performance. Instead of relying on increasingly complex architectures or multi-stage training schemes, TC-AE addresses this challenge from the perspective of the token space, the key bridge between pixels and image latents, through two complementary innovations: Firstly, we study token number scaling by adjusting the patch size in ViT under a fixed latent budget, and identify aggressive token-to-latent compression as the key factor that limits effective scaling. To address this issue, we decompose token-to-latent compression into two stages, reducing structural information loss and enabling effective token number scaling for generation. Secondly, to further mitigate latent representation collapse, we enhance the semantic structure of image tokens via joint self-supervised training, leading to more generative-friendly latents. With these designs, TC-AE achieves substantially improved reconstruction and generative performance under deep compression. We hope our research will advance ViT-based tokenizer for visual generation.

Rui Jiang, Xinghe Fu, Guangcong Zheng et al.

The rapid advancement of pretrained text-driven diffusion models has significantly enriched applications in image generation and editing. However, as the demand for personalized content editing increases, new challenges emerge especially when dealing with arbitrary objects and complex scenes. Existing methods usually mistakes mask as the object shape prior, which struggle to achieve a seamless integration result. The mostly used inversion noise initialization also hinders the identity consistency towards the target object. To address these challenges, we propose a novel training-free framework that formulates personalized content editing as the optimization of edited images in the latent space, using diffusion models as the energy function guidance conditioned by reference text-image pairs. A coarse-to-fine strategy is proposed that employs text energy guidance at the early stage to achieve a natural transition toward the target class and uses point-to-point feature-level image energy guidance to perform fine-grained appearance alignment with the target object. Additionally, we introduce the latent space content composition to enhance overall identity consistency with the target. Extensive experiments demonstrate that our method excels in object replacement even with a large domain gap, highlighting its potential for high-quality, personalized image editing.

Teng Li, Hulya Seferoglu

Distributed inference techniques can be broadly classified into data-distributed and model-distributed schemes. In data-distributed inference (DDI), each worker carries the entire Machine Learning (ML) model but processes only a subset of the data. However, feeding the data to workers results in high communication costs, especially when the data is large. An emerging paradigm is model-distributed inference (MDI), where each worker carries only a subset of ML layers. In MDI, a source device that has data processes a few layers of ML model and sends the output to a neighboring device, i.e., offloads the rest of the layers. This process ends when all layers are processed in a distributed manner. In this paper, we investigate the design and development of MDI when multiple data sources co-exist. We consider that each data source has a different importance and, hence, a priority. We formulate and solve a priority-aware model allocation optimization problem. Based on the structure of the optimal solution, we design a practical Priority-Aware Model- Distributed Inference (PA-MDI) algorithm that determines model allocation and distribution over devices by taking into account the priorities of different sources. Experiments were conducted on a real-life testbed of NVIDIA Jetson Xavier and Nano edge devices as well as in the Colosseum testbed with ResNet-50, ResNet- 56, and GPT-2 models. The experimental results show that PA-MDI performs priority-aware model allocation successfully while reducing the inference time as compared to baselines.

Sicong Liu, Weiye Wu, Xiangrui Xu et al.

Foundation models have reshaped AI by unifying fragmented architectures into scalable backbones with multimodal reasoning and contextual adaptation. In parallel, the long-standing notion of AI agents, defined by the sensing-decision-action loop, is entering a new paradigm: with FMs as their cognitive core, agents transcend rule-based behaviors to achieve autonomy, generalization, and self-reflection. This dual shift is reinforced by real-world demands such as autonomous driving, robotics, virtual assistants, and GUI agents, as well as ecosystem advances in embedded hardware, edge computing, mobile deployment platforms, and communication protocols that together enable large-scale deployment. Yet this convergence collides with reality: while applications demand long-term adaptability and real-time interaction, mobile and edge deployments remain constrained by memory, energy, bandwidth, and latency. This creates a fundamental tension between the growing complexity of FMs and the limited resources of deployment environments. This survey provides the first systematic characterization of adaptive, resource-efficient agentic AI systems. We summarize enabling techniques into elastic inference, test-time adaptation, dynamic multimodal integration, and agentic AI applications, and identify open challenges in balancing accuracy-latency-communication trade-offs and sustaining robustness under distribution shifts. We further highlight future opportunities in algorithm-system co-design, cognitive adaptation, and collaborative edge deployment. By mapping FM structures, cognition, and hardware resources, this work establishes a unified perspective toward scalable, adaptive, and resource-efficient agentic AI. We believe this survey can help readers to understand the connections between enabling technologies while promoting further discussions on the fusion of agentic intelligence and intelligent agents.

Yangzhou Jiang, Yinxin Lin, Yaoming Wang et al.

Appearance-based supervised methods with full-face image input have made tremendous advances in recent gaze estimation tasks. However, intensive human annotation requirement inhibits current methods from achieving industrial level accuracy and robustness. Although current unsupervised pre-training frameworks have achieved success in many image recognition tasks, due to the deep coupling between facial and eye features, such frameworks are still deficient in extracting useful gaze features from full-face. To alleviate above limitations, this work proposes a novel unsupervised/self-supervised gaze pre-training framework, which forces the full-face branch to learn a low dimensional gaze embedding without gaze annotations, through collaborative feature contrast and squeeze modules. In the heart of this framework is an alternating eye-attended/unattended masking training scheme, which squeezes gaze-related information from full-face branch into an eye-masked auto-encoder through an injection bottleneck design that successfully encourages the model to pays more attention to gaze direction rather than facial textures only, while still adopting the eye self-reconstruction objective. In the same time, a novel eye/gaze-related information contrastive loss has been designed to further boost the learned representation by forcing the model to focus on eye-centered regions. Extensive experimental results on several gaze benchmarks demonstrate that the proposed scheme achieves superior performances over unsupervised state-of-the-art.

Teng Li, Siwei Yin, Runze Yu et al.

To fight against infectious diseases (e.g., SARS, COVID-19, Ebola, etc.), government agencies, technology companies and health institutes have launched various contact tracing approaches to identify and notify the people exposed to infection sources. However, existing tracing approaches can lead to severe privacy and security concerns, thereby preventing their secure and widespread use among communities. To tackle these problems, this paper proposes CoAvoid, a decentralized, privacy-preserved contact tracing system that features good dependability and usability. CoAvoid leverages the Google/Apple Exposure Notification (GAEN) API to achieve decent device compatibility and operating efficiency. It utilizes GPS along with Bluetooth Low Energy (BLE) to dependably verify user information. In addition, to enhance privacy protection, CoAvoid applies fuzzification and obfuscation measures to shelter sensitive data, making both servers and users agnostic to information of both low and high-risk populations. The evaluation demonstrates good efficacy and security of CoAvoid. Compared with four state-of-art contact tracing applications, CoAvoid can reduce upload data by at least 90% and simultaneously resist wormhole and replay attacks in various scenarios.

Shen Li, Yanli Zhao, Rohan Varma et al.

This paper presents the design, implementation, and evaluation of the PyTorch distributed data parallel module. PyTorch is a widely-adopted scientific computing package used in deep learning research and applications. Recent advances in deep learning argue for the value of large datasets and large models, which necessitates the ability to scale out model training to more computational resources. Data parallelism has emerged as a popular solution for distributed training thanks to its straightforward principle and broad applicability. In general, the technique of distributed data parallelism replicates the model on every computational resource to generate gradients independently and then communicates those gradients at each iteration to keep model replicas consistent. Despite the conceptual simplicity of the technique, the subtle dependencies between computation and communication make it non-trivial to optimize the distributed training efficiency. As of v1.5, PyTorch natively provides several techniques to accelerate distributed data parallel, including bucketing gradients, overlapping computation with communication, and skipping gradient synchronization. Evaluations show that, when configured appropriately, the PyTorch distributed data parallel module attains near-linear scalability using 256 GPUs.

Minghao Xu, Jian Zhang, Bingbing Ni et al.

Recent works on domain adaptation reveal the effectiveness of adversarial learning on filling the discrepancy between source and target domains. However, two common limitations exist in current adversarial-learning-based methods. First, samples from two domains alone are not sufficient to ensure domain-invariance at most part of latent space. Second, the domain discriminator involved in these methods can only judge real or fake with the guidance of hard label, while it is more reasonable to use soft scores to evaluate the generated images or features, i.e., to fully utilize the inter-domain information. In this paper, we present adversarial domain adaptation with domain mixup (DM-ADA), which guarantees domain-invariance in a more continuous latent space and guides the domain discriminator in judging samples' difference relative to source and target domains. Domain mixup is jointly conducted on pixel and feature level to improve the robustness of models. Extensive experiments prove that the proposed approach can achieve superior performance on tasks with various degrees of domain shift and data complexity.

Teng Li, Chaoqun Wang, Max Q. -H. Meng et al.

Unmanned Aerial Vehicles (UAVs) have been implemented for environmental monitoring by using their capabilities of mobile sensing, autonomous navigation, and remote operation. However, in real-world applications, the limitations of on-board resources (e.g., power supply) of UAVs will constrain the coverage of the monitored area and the number of the acquired samples, which will hinder the performance of field estimation and mapping. Therefore, the issue of constrained resources calls for an efficient sampling planner to schedule UAV-based sensing tasks in environmental monitoring. This paper presents a mission planner of coverage sampling and path planning for a UAV-enabled mobile sensor to effectively explore and map an unknown environment that is modeled as a random field. The proposed planner can generate a coverage path with an optimal coverage density for exploratory sampling, and the associated energy cost is subjected to a power supply constraint. The performance of the developed framework is evaluated and compared with the existing state-of-the-art algorithms, using a real-world dataset that is collected from an environmental monitoring program as well as physical field experiments. The experimental results illustrate the reliability and accuracy of the presented coverage sampling planner in a prior survey for environmental exploration and field mapping.

Hyo-Sang Shin, Teng Li, Pau Segui-Gasco

This paper addresses the task allocation problem for multi-robot systems. The main issue with the task allocation problem is inherent complexity that makes finding an optimal solution within a reasonable time almost impossible. To hand the issue, this paper develops a task allocation algorithm that can be decentralised by leveraging the submodularity concepts and sampling process. The theoretical analysis reveals that the proposed algorithm can provide approximation guarantee of $1/2$ for the monotone submodular case and $1/4$ for the non-monotone submodular case in average sense with polynomial time complexity. To examine the performance of the proposed algorithm and validate the theoretical analysis results, we design a task allocation problem and perform numerical simulations. The simulation results confirm that the proposed algorithm achieves solution quality, which is comparable to a state-of-the-art algorithm in the monotone case, and much better quality in the non-monotone case with significantly less computational complexity.

Chaoqun Wang, Delong Zhu, Teng Li et al.

In this paper, we propose an integrated framework for the autonomous robotic exploration in indoor environments. Specially, we present a hybrid map, named Semantic Road Map (SRM), to represent the topological structure of the explored environment and facilitate decision-making in the exploration. The SRM is built incrementally along with the exploration process. It is a graph structure with collision-free nodes and edges that are generated within the sensor coverage. Moreover, each node has a semantic label and the expected information gain at that location. Based on the concise SRM, we present a novel and effective decision-making model to determine the next-best-target (NBT) during the exploration. The model concerns the semantic information, the information gain, and the path cost to the target location. We use the nodes of SRM to represent the candidate targets, which enables the target evaluation to be performed directly on the SRM. With the SRM, both the information gain of a node and the path cost to the node can be obtained efficiently. Besides, we adopt the cross-entropy method to optimize the path to make it more informative. We conduct experimental studies in both simulated and real-world environments, which demonstrate the effectiveness of the proposed method.

Wei Tang, Gui Li, Xinyuan Bao et al.

To synthesize high-quality person images with arbitrary poses is challenging. In this paper, we propose a novel Multi-scale Conditional Generative Adversarial Networks (MsCGAN), aiming to convert the input conditional person image to a synthetic image of any given target pose, whose appearance and the texture are consistent with the input image. MsCGAN is a multi-scale adversarial network consisting of two generators and two discriminators. One generator transforms the conditional person image into a coarse image of the target pose globally, and the other is to enhance the detailed quality of the synthetic person image through a local reinforcement network. The outputs of the two generators are then merged into a synthetic, discriminant and high-resolution image. On the other hand, the synthetic image is downsampled to multiple resolutions as the input to multi-scale discriminator networks. The proposed multi-scale generators and discriminators handling different levels of visual features can benefit to synthesizing high-resolution person images with realistic appearance and texture. Experiments are conducted on the Market-1501 and DeepFashion datasets to evaluate the proposed model, and both qualitative and quantitative results demonstrate the superior performance of the proposed MsCGAN.

Teng Li, Zhiyuan Xu, Jian Tang et al.

In this paper, we focus on general-purpose Distributed Stream Data Processing Systems (DSDPSs), which deal with processing of unbounded streams of continuous data at scale distributedly in real or near-real time. A fundamental problem in a DSDPS is the scheduling problem with the objective of minimizing average end-to-end tuple processing time. A widely-used solution is to distribute workload evenly over machines in the cluster in a round-robin manner, which is obviously not efficient due to lack of consideration for communication delay. Model-based approaches do not work well either due to the high complexity of the system environment. We aim to develop a novel model-free approach that can learn to well control a DSDPS from its experience rather than accurate and mathematically solvable system models, just as a human learns a skill (such as cooking, driving, swimming, etc). Specifically, we, for the first time, propose to leverage emerging Deep Reinforcement Learning (DRL) for enabling model-free control in DSDPSs; and present design, implementation and evaluation of a novel and highly effective DRL-based control framework, which minimizes average end-to-end tuple processing time by jointly learning the system environment via collecting very limited runtime statistics data and making decisions under the guidance of powerful Deep Neural Networks. To validate and evaluate the proposed framework, we implemented it based on a widely-used DSDPS, Apache Storm, and tested it with three representative applications. Extensive experimental results show 1) Compared to Storm's default scheduler and the state-of-the-art model-based method, the proposed framework reduces average tuple processing by 33.5% and 14.0% respectively on average. 2) The proposed framework can quickly reach a good scheduling solution during online learning, which justifies its practicability for online control in DSDPSs.

Chaoqun Wang, Lili Meng, Sizhen She et al.

Robots are increasingly operating in indoor environments designed for and shared with people. However, robots working safely and autonomously in uneven and unstructured environments still face great challenges. Many modern indoor environments are designed with wheelchair accessibility in mind. This presents an opportunity for wheeled robots to navigate through sloped areas while avoiding staircases. In this paper, we present an integrated software and hardware system for autonomous mobile robot navigation in uneven and unstructured indoor environments. This modular and reusable software framework incorporates capabilities of perception and navigation. Our robot first builds a 3D OctoMap representation for the uneven environment with the 3D mapping using wheel odometry, 2D laser and RGB-D data. Then we project multilayer 2D occupancy maps from OctoMap to generate the the traversable map based on layer differences. The safe traversable map serves as the input for efficient autonomous navigation. Furthermore, we employ a variable step size Rapidly Exploring Random Trees that could adjust the step size automatically, eliminating tuning step sizes according to environments. We conduct extensive experiments in simulation and real-world, demonstrating the efficacy and efficiency of our system.

Xiangyun Zhao, Xiaodan Liang, Luoqi Liu et al.

Objective functions for training of deep networks for face-related recognition tasks, such as facial expression recognition (FER), usually consider each sample independently. In this work, we present a novel peak-piloted deep network (PPDN) that uses a sample with peak expression (easy sample) to supervise the intermediate feature responses for a sample of non-peak expression (hard sample) of the same type and from the same subject. The expression evolving process from non-peak expression to peak expression can thus be implicitly embedded in the network to achieve the invariance to expression intensities. A special purpose back-propagation procedure, peak gradient suppression (PGS), is proposed for network training. It drives the intermediate-layer feature responses of non-peak expression samples towards those of the corresponding peak expression samples, while avoiding the inverse. This avoids degrading the recognition capability for samples of peak expression due to interference from their non-peak expression counterparts. Extensive comparisons on two popular FER datasets, Oulu-CASIA and CK+, demonstrate the superiority of the PPDN over state-ofthe-art FER methods, as well as the advantages of both the network structure and the optimization strategy. Moreover, it is shown that PPDN is a general architecture, extensible to other tasks by proper definition of peak and non-peak samples. This is validated by experiments that show state-of-the-art performance on pose-invariant face recognition, using the Multi-PIE dataset.

Teng Li, Huan Chang, Meng Wang et al.

Automated scene analysis has been a topic of great interest in computer vision and cognitive science. Recently, with the growth of crowd phenomena in the real world, crowded scene analysis has attracted much attention. However, the visual occlusions and ambiguities in crowded scenes, as well as the complex behaviors and scene semantics, make the analysis a challenging task. In the past few years, an increasing number of works on crowded scene analysis have been reported, covering different aspects including crowd motion pattern learning, crowd behavior and activity analysis, and anomaly detection in crowds. This paper surveys the state-of-the-art techniques on this topic. We first provide the background knowledge and the available features related to crowded scenes. Then, existing models, popular algorithms, evaluation protocols, as well as system performance are provided corresponding to different aspects of crowded scene analysis. We also outline the available datasets for performance evaluation. Finally, some research problems and promising future directions are presented with discussions.