Fei Hu

Mingqiang Wei, Zeyong Wei, Haoran Zhou et al.

Convolution on 3D point clouds is widely researched yet far from perfect in geometric deep learning. The traditional wisdom of convolution characterises feature correspondences indistinguishably among 3D points, arising an intrinsic limitation of poor distinctive feature learning. In this paper, we propose Adaptive Graph Convolution (AGConv) for wide applications of point cloud analysis. AGConv generates adaptive kernels for points according to their dynamically learned features. Compared with the solution of using fixed/isotropic kernels, AGConv improves the flexibility of point cloud convolutions, effectively and precisely capturing the diverse relations between points from different semantic parts. Unlike the popular attentional weight schemes, AGConv implements the adaptiveness inside the convolution operation instead of simply assigning different weights to the neighboring points. Extensive evaluations clearly show that our method outperforms state-of-the-arts of point cloud classification and segmentation on various benchmark datasets.Meanwhile, AGConv can flexibly serve more point cloud analysis approaches to boost their performance. To validate its flexibility and effectiveness, we explore AGConv-based paradigms of completion, denoising, upsampling, registration and circle extraction, which are comparable or even superior to their competitors. Our code is available at https://github.com/hrzhou2/AdaptConv-master.

Gongjin Lan, Yu Wu, Fei Hu et al.

Human pose estimation (HPE) has attracted a significant amount of attention from the computer vision community in the past decades. Moreover, HPE has been applied to various domains, such as human-computer interaction, sports analysis, and human tracking via images and videos. Recently, deep learning-based approaches have shown state-of-the-art performance in HPE-based applications. Although deep learning-based approaches have achieved remarkable performance in HPE, a comprehensive review of deep learning-based HPE methods remains lacking in the literature. In this article, we provide an up-to-date and in-depth overview of the deep learning approaches in vision-based HPE. We summarize these methods of 2-D and 3-D HPE, and their applications, discuss the challenges and the research trends through bibliometrics, and provide insightful recommendations for future research. This article provides a meaningful overview as introductory material for beginners to deep learning-based HPE, as well as supplementary material for advanced researchers.

Fei Hu, Honghua Chen, Xuequan Lu et al.

How will you repair a physical object with large missings? You may first recover its global yet coarse shape and stepwise increase its local details. We are motivated to imitate the above physical repair procedure to address the point cloud completion task. We propose a novel stepwise point cloud completion network (SPCNet) for various 3D models with large missings. SPCNet has a hierarchical bottom-to-up network architecture. It fulfills shape completion in an iterative manner, which 1) first infers the global feature of the coarse result; 2) then infers the local feature with the aid of global feature; and 3) finally infers the detailed result with the help of local feature and coarse result. Beyond the wisdom of simulating the physical repair, we newly design a cycle loss %based training strategy to enhance the generalization and robustness of SPCNet. Extensive experiments clearly show the superiority of our SPCNet over the state-of-the-art methods on 3D point clouds with large missings.

Cheng Guo, Fei Hu, Yan Hu

Passive millimeter-wave (PMMW) is a significant potential technique for human security screening. Several popular object detection networks have been used for PMMW images. However, restricted by the low resolution and high noise of PMMW images, PMMW hidden object detection based on deep learning usually suffers from low accuracy and low classification confidence. To tackle the above problems, this paper proposes a Task-Aligned Detection Transformer network, named PMMW-DETR. In the first stage, a Denoising Coarse-to-Fine Transformer (DCFT) backbone is designed to extract long- and short-range features in the different scales. In the second stage, we propose the Query Selection module to introduce learned spatial features into the network as prior knowledge, which enhances the semantic perception capability of the network. In the third stage, aiming to improve the classification performance, we perform a Task-Aligned Dual-Head block to decouple the classification and regression tasks. Based on our self-developed PMMW security screening dataset, experimental results including comparison with State-Of-The-Art (SOTA) methods and ablation study demonstrate that the PMMW-DETR obtains higher accuracy and classification confidence than previous works, and exhibits robustness to the PMMW images of low quality.

Liang Hu, Jianpeng Jiao, Jiashuo Liu et al.

Search has emerged as core infrastructure for LLM-based agents and is widely viewed as critical on the path toward more general intelligence. Finance is a particularly demanding proving ground: analysts routinely conduct complex, multi-step searches over time-sensitive, domain-specific data, making it ideal for assessing both search proficiency and knowledge-grounded reasoning. Yet no existing open financial datasets evaluate data searching capability of end-to-end agents, largely because constructing realistic, complicated tasks requires deep financial expertise and time-sensitive data is hard to evaluate. We present FinSearchComp, the first fully open-source agent benchmark for realistic, open-domain financial search and reasoning. FinSearchComp comprises three tasks -- Time-Sensitive Data Fetching, Simple Historical Lookup, and Complex Historical Investigation -- closely reproduce real-world financial analyst workflows. To ensure difficulty and reliability, we engage 70 professional financial experts for annotation and implement a rigorous multi-stage quality-assurance pipeline. The benchmark includes 635 questions spanning global and Greater China markets, and we evaluate 21 models (products) on it. Grok 4 (web) tops the global subset, approaching expert-level accuracy. DouBao (web) leads on the Greater China subset. Experimental analyses show that equipping agents with web search and financial plugins substantially improves results on FinSearchComp, and the country origin of models and tools impact performance significantly.By aligning with realistic analyst tasks and providing end-to-end evaluation, FinSearchComp offers a professional, high-difficulty testbed for complex financial search and reasoning.

Zhicheng Zhou, Hailong Chen, Kunhua Li et al.

Federated Learning (FL) since proposed has been applied in many fields, such as credit assessment, medical, etc. Because of the difference in the network or computing resource, the clients may not update their gradients at the same time that may take a lot of time to wait or idle. That's why Asynchronous Federated Learning (AFL) method is needed. The main bottleneck in AFL is communication. How to find a balance between the model performance and the communication cost is a challenge in AFL. This paper proposed a novel AFL framework VAFL. And we verified the performance of the algorithm through sufficient experiments. The experiments show that VAFL can reduce the communication times about 51.02\% with 48.23\% average communication compression rate and allow the model to be converged faster. The code is available at \url{https://github.com/RobAI-Lab/VAFL}

Yu Yan, Fei Hu, Jiusheng Chen et al.

Transformer-based models have made tremendous impacts in natural language generation. However the inference speed is a bottleneck due to large model size and intensive computing involved in auto-regressive decoding process. We develop FastSeq framework to accelerate sequence generation without accuracy loss. The proposed optimization techniques include an attention cache optimization, an efficient algorithm for detecting repeated n-grams, and an asynchronous generation pipeline with parallel I/O. These optimizations are general enough to be applicable to Transformer-based models (e.g., T5, GPT2, and UniLM). Our benchmark results on a set of widely used and diverse models demonstrate 4-9x inference speed gain. Additionally, FastSeq is easy to use with a simple one-line code change. The source code is available at https://github.com/microsoft/fastseq.

Alex Bozarth, Brendan Dwyer, Fei Hu et al.

A recent trend observed in traditionally challenging fields such as computer vision and natural language processing has been the significant performance gains shown by deep learning (DL). In many different research fields, DL models have been evolving rapidly and become ubiquitous. Despite researchers' excitement, unfortunately, most software developers are not DL experts and oftentimes have a difficult time following the booming DL research outputs. As a result, it usually takes a significant amount of time for the latest superior DL models to prevail in industry. This issue is further exacerbated by the common use of sundry incompatible DL programming frameworks, such as Tensorflow, PyTorch, Theano, etc. To address this issue, we propose a system, called Model Asset Exchange (MAX), that avails developers of easy access to state-of-the-art DL models. Regardless of the underlying DL programming frameworks, it provides an open source Python library (called the MAX framework) that wraps DL models and unifies programming interfaces with our standardized RESTful APIs. These RESTful APIs enable developers to exploit the wrapped DL models for inference tasks without the need to fully understand different DL programming frameworks. Using MAX, we have wrapped and open-sourced more than 30 state-of-the-art DL models from various research fields, including computer vision, natural language processing and signal processing, etc. In the end, we selectively demonstrate two web applications that are built on top of MAX, as well as the process of adding a DL model to MAX.

Tianyu Fu, Anyang Su, Chenxu Zhao et al.

Graphical user interfaces (GUIs) are the primary medium for human-computer interaction, yet automating GUI interactions remains challenging due to the complexity of visual elements, dynamic environments, and the need for multi-step reasoning. Existing methods based on vision-language models (VLMs) often suffer from limited resolution, domain mismatch, and insufficient sequential decisionmaking capability. To address these issues, we propose Mano, a robust GUI agent built upon a multi-modal foundation model pre-trained on extensive web and computer system data. Our approach integrates a novel simulated environment for high-fidelity data generation, a three-stage training pipeline (supervised fine-tuning, offline reinforcement learning, and online reinforcement learning), and a verification module for error recovery. Mano demonstrates state-of-the-art performance on multiple GUI benchmarks, including Mind2Web and OSWorld, achieving significant improvements in success rate and operational accuracy. Our work provides new insights into the effective integration of reinforcement learning with VLMs for practical GUI agent deployment, highlighting the importance of domain-specific data, iterative training, and holistic reward design.

Qinyan Zhang, Xinping Lei, Ruijie Miao et al.

Large Language Models (LLMs) achieve strong performance on diverse tasks but often exhibit cognitive inertia, struggling to follow instructions that conflict with the standardized patterns learned during supervised fine-tuning (SFT). To evaluate this limitation, we propose Inverse IFEval, a benchmark that measures models Counter-intuitive Abilitytheir capacity to override training-induced biases and comply with adversarial instructions. Inverse IFEval introduces eight types of such challenges, including Question Correction, Intentional Textual Flaws, Code without Comments, and Counterfactual Answering. Using a human-in-the-loop pipeline, we construct a dataset of 1012 high-quality Chinese and English questions across 23 domains, evaluated under an optimized LLM-as-a-Judge framework. Experiments on existing leading LLMs demonstrate the necessity of our proposed Inverse IFEval benchmark. Our findings emphasize that future alignment efforts should not only pursue fluency and factual correctness but also account for adaptability under unconventional contexts. We hope that Inverse IFEval serves as both a diagnostic tool and a foundation for developing methods that mitigate cognitive inertia, reduce overfitting to narrow patterns, and ultimately enhance the instruction-following reliability of LLMs in diverse and unpredictable real-world scenarios.

Jingzhe Ding, Shengda Long, Changxin Pu et al.

Recent advances in coding agents suggest rapid progress toward autonomous software development, yet existing benchmarks fail to rigorously evaluate the long-horizon capabilities required to build complete software systems. Most prior evaluations focus on localized code generation, scaffolded completion, or short-term repair tasks, leaving open the question of whether agents can sustain coherent reasoning, planning, and execution over the extended horizons demanded by real-world repository construction. To address this gap, we present NL2Repo Bench, a benchmark explicitly designed to evaluate the long-horizon repository generation ability of coding agents. Given only a single natural-language requirements document and an empty workspace, agents must autonomously design the architecture, manage dependencies, implement multi-module logic, and produce a fully installable Python library. Our experiments across state-of-the-art open- and closed-source models reveal that long-horizon repository generation remains largely unsolved: even the strongest agents achieve below 40% average test pass rates and rarely complete an entire repository correctly. Detailed analysis uncovers fundamental long-horizon failure modes, including premature termination, loss of global coherence, fragile cross-file dependencies, and inadequate planning over hundreds of interaction steps. NL2Repo Bench establishes a rigorous, verifiable testbed for measuring sustained agentic competence and highlights long-horizon reasoning as a central bottleneck for the next generation of autonomous coding agents.

Xinxin Dong, Baoyun Peng, Haokai Ma et al.

Video Question Answering (VideoQA) requires identifying sparse critical moments in long videos and reasoning about their causal relationships to answer semantically complex questions. While recent advances in multimodal learning have improved alignment and fusion, current approaches remain limited by two prevalent but fundamentally flawed strategies: (1) task-agnostic sampling indiscriminately processes all frames, overwhelming key events with irrelevant content; and (2) heuristic retrieval captures superficial patterns but misses causal-temporal structures needed for complex reasoning. To address these challenges, we introduce LeAdQA, an innovative approach that bridges these gaps through synergizing causal-aware query refinement with fine-grained visual grounding. Our method first leverages LLMs to reformulate question-option pairs, resolving causal ambiguities and sharpening temporal focus. These refined queries subsequently direct a temporal grounding model to precisely retrieve the most salient segments, complemented by an adaptive fusion mechanism dynamically integrating the evidence to maximize relevance. The integrated visual-textual cues are then processed by an MLLM to generate accurate, contextually-grounded answers. Experiments on NExT-QA, IntentQA, and NExT-GQA demonstrate that our method's precise visual grounding substantially enhances the understanding of video-question relationships, achieving state-of-the-art (SOTA) performance on complex reasoning tasks while maintaining computational efficiency.

Jiyao Zhang, Weiyao Huang, Bo Peng et al.

6D Object Pose Estimation is a crucial yet challenging task in computer vision, suffering from a significant lack of large-scale datasets. This scarcity impedes comprehensive evaluation of model performance, limiting research advancements. Furthermore, the restricted number of available instances or categories curtails its applications. To address these issues, this paper introduces Omni6DPose, a substantial dataset characterized by its diversity in object categories, large scale, and variety in object materials. Omni6DPose is divided into three main components: ROPE (Real 6D Object Pose Estimation Dataset), which includes 332K images annotated with over 1.5M annotations across 581 instances in 149 categories; SOPE(Simulated 6D Object Pose Estimation Dataset), consisting of 475K images created in a mixed reality setting with depth simulation, annotated with over 5M annotations across 4162 instances in the same 149 categories; and the manually aligned real scanned objects used in both ROPE and SOPE. Omni6DPose is inherently challenging due to the substantial variations and ambiguities. To address this challenge, we introduce GenPose++, an enhanced version of the SOTA category-level pose estimation framework, incorporating two pivotal improvements: Semantic-aware feature extraction and Clustering-based aggregation. Moreover, we provide a comprehensive benchmarking analysis to evaluate the performance of previous methods on this large-scale dataset in the realms of 6D object pose estimation and pose tracking.

Chao Jiang, Canchen Jiang, Dongwei Chen et al.

Densely connected convolutional networks (DenseNet) behave well in image processing. However, for regression tasks, convolutional DenseNet may lose essential information from independent input features. To tackle this issue, we propose a novel DenseNet regression model where convolution and pooling layers are replaced by fully connected layers and the original concatenation shortcuts are maintained to reuse the feature. To investigate the effects of depth and input dimension of proposed model, careful validations are performed by extensive numerical simulation. The results give an optimal depth (19) and recommend a limited input dimension (under 200). Furthermore, compared with the baseline models including support vector regression, decision tree regression, and residual regression, our proposed model with the optimal depth performs best. Ultimately, DenseNet regression is applied to predict relative humidity, and the outcome shows a high correlation (0.91) with observations, which indicates that our model could advance environmental data analysis.

Xiao-Wei Tang, Xin-Lin Huang, Fei Hu

The explosive demands for high quality mobile video services have caused heavy overload to the existing cellular networks. Although the small cell has been proposed to alleviate such a problem, the network operators may not be interested in deploying numerous base stations (BSs) due to expensive infrastructure construction and maintenance. The unmanned aerial vehicles (UAVs) can provide the low-cost and quick deployment, which can support high-quality line-of-sight communications and have become promising mobile BSs. In this paper, we propose a quality-of-experience (QoE)-driven UAV-enabled pseudo-analog wireless video broadcast scheme, which provides mobile video broadcast services for ground users (GUs). Due to limited energy available in UAV, the aim of the proposed scheme is to maximize the minimum peak signal-to-noise ratio (PSNR) of GUs' video reconstruction quality by jointly optimizing the transmission power allocation strategy and the UAV trajectory. Firstly, the reconstructed video quality at GUs is defined under the constraints of the UAV's total energy and motion mechanism, and the proposed scheme is formulated as a complex non-convex optimization problem. Then, the optimization problem is simplified to obtain a tractable suboptimal solution with the help of the block coordinate descent model and the successive convex approximation model. Finally, the experimental results are presented to show the effectiveness of the proposed scheme. Specifically, the proposed scheme can achieve over 1.6dB PSNR gains in terms of GUs' minimum PSNR, compared with the state-of-the-art schemes, e.g., DVB, SoftCast, and SharpCast.

Xiao-Wei Tang, Xin-Lin Huang, Fei Hu et al.

Recently, pseudo analog transmission has gained increasing attentions due to its ability to alleviate the cliff effect in video multicast scenarios. The existing pseudo analog systems are sorely optimized under the minimum mean squared error criterion without taking the perceptual video quality into consideration. In this paper, we propose a human-perception-based pseudo analog video transmission system named ROIC-Cast, which aims to intelligently enhance the transmission quality of the region-of-interest (ROI) parts. Firstly, the classic deep learning based saliency detection algorithm is adopted to decompose the continuous video sequences into ROI and non-ROI blocks. Secondly, an effective compression method is used to reduce the data amount of side information generated by the ROI extraction module. Then, the power allocation scheme is formulated as a convex problem, and the optimal transmission power for both ROI and non-ROI blocks is derived in a closed form. Finally, the simulations are conducted to validate the proposed system by comparing with a few of existing systems, e.g., KMV-Cast, SoftCast, and DAC-RAN. The proposed ROIC-Cast can achieve over 4.1dB peak signal- to-noise ratio gains of ROI compared with other systems, given the channel signal-to-noise ratio as -5dB, 0dB, 5dB, and 10dB, respectively. This significant performance improvement is due to the automatic ROI extraction, high-efficiency data compression as well as adaptive power allocation.

Fei Hu, Wei Liu, Ajmal Saeed Mian et al.

Although response generation (RG) diversification for single-turn dialogs has been well developed, it is less investigated for natural multi-turn conversations. Besides, past work focused on diversifying responses without considering topic coherence to the context, producing uninformative replies. In this paper, we propose the Topic-coherent Hierarchical Recurrent Encoder-Decoder model (THRED) to diversify the generated responses without deviating the contextual topics for multi-turn conversations. In overall, we build a sequence-to-sequence net (Seq2Seq) to model multi-turn conversations. And then we resort to the latent Variable Hierarchical Recurrent Encoder-Decoder model (VHRED) to learn global contextual distribution of dialogs. Besides, we construct a dense topic matrix which implies word-level correlations of the conversation corpora. The topic matrix is used to learn local topic distribution of the contextual utterances. By incorporating both the global contextual distribution and the local topic distribution, THRED produces both diversified and topic-coherent replies. In addition, we propose an explicit metric (\emph{TopicDiv}) to measure the topic divergence between the post and generated response, and we also propose an overall metric combining the diversification metric (\emph{Distinct}) and \emph{TopicDiv}. We evaluate our model comparing with three baselines (Seq2Seq, HRED and VHRED) on two real-world corpora, respectively, and demonstrate its outstanding performance in both diversification and topic coherence.

Xiaofei Xu, Ke Deng, Fei Hu et al.

Many words have evolved in meaning as a result of cultural and social change. Understanding such changes is crucial for modelling language and cultural evolution. Low-dimensional embedding methods have shown promise in detecting words' meaning change by encoding them into dense vectors. However, when exploring semantic change of words over time, these methods require the alignment of word embeddings across different time periods. This process is computationally expensive, prohibitively time consuming and suffering from contextual variability. In this paper, we propose a new and scalable method for encoding words from different time periods into one dense vector space. This can greatly improve performance when it comes to identifying words that have changed in meaning over time. We evaluated our method on dataset from Google Books N-gram. Our method outperformed three other popular methods in terms of the number of words correctly identified to have changed in meaning. Additionally, we provide an intuitive visualization of the semantic evolution of some words extracted by our method

Fei Hu, Xiaofei Xu, Tao Peng et al.

Based on Restricted Boltzmann Machines (RBMs), an improved pseudo-stochastic sequential cipher generator is proposed. It is effective and efficient because of the two advantages: this generator includes a stochastic neural network that can perform the calculation in parallel, that is to say, all elements are calculated simultaneously; unlimited number of sequential ciphers can be generated simultaneously for multiple encryption schemas. The periodicity and the correlation of the output sequential ciphers meet the requirements for the design of encrypting sequential data. In the experiment, the generated sequential cipher is used to encrypt the image, and better performance is achieved in terms of the key space analysis, the correlation analysis, the sensitivity analysis and the differential attack. The experimental result is promising that could promote the development of image protection in computer security.

Fei Hu, Changjiu Pu, Haowei Gao et al.

Stacked Auto-Encoder (SAE) is a kind of deep learning algorithm for unsupervised learning. Which has multi layers that project the vector representation of input data into a lower vector space. These projection vectors are dense representations of the input data. As a result, SAE can be used for image compression. Using chaotic logistic map, the compression ones can further be encrypted. In this study, an application of image compression and encryption is suggested using SAE and chaotic logistic map. Experiments show that this application is feasible and effective. It can be used for image transmission and image protection on internet simultaneously.