Ning Xie

Pedro Miguel Sánchez Sánchez, Alberto Huertas Celdrán, Ning Xie et al.

The rapid expansion of the Internet of Things (IoT) and Edge Computing has presented challenges for centralized Machine and Deep Learning (ML/DL) methods due to the presence of distributed data silos that hold sensitive information. To address concerns regarding data privacy, collaborative and privacy-preserving ML/DL techniques like Federated Learning (FL) have emerged. However, ensuring data privacy and performance alone is insufficient since there is a growing need to establish trust in model predictions. Existing literature has proposed various approaches on trustworthy ML/DL (excluding data privacy), identifying robustness, fairness, explainability, and accountability as important pillars. Nevertheless, further research is required to identify trustworthiness pillars and evaluation metrics specifically relevant to FL models, as well as to develop solutions that can compute the trustworthiness level of FL models. This work examines the existing requirements for evaluating trustworthiness in FL and introduces a comprehensive taxonomy consisting of six pillars (privacy, robustness, fairness, explainability, accountability, and federation), along with over 30 metrics for computing the trustworthiness of FL models. Subsequently, an algorithm named FederatedTrust is designed based on the pillars and metrics identified in the taxonomy to compute the trustworthiness score of FL models. A prototype of FederatedTrust is implemented and integrated into the learning process of FederatedScope, a well-established FL framework. Finally, five experiments are conducted using different configurations of FederatedScope to demonstrate the utility of FederatedTrust in computing the trustworthiness of FL models. Three experiments employ the FEMNIST dataset, and two utilize the N-BaIoT dataset considering a real-world IoT security use case.

Yuxia Wang, Shimin Tao, Ning Xie et al.

Despite the subjective nature of semantic textual similarity (STS) and pervasive disagreements in STS annotation, existing benchmarks have used averaged human ratings as the gold standard. Averaging masks the true distribution of human opinions on examples of low agreement, and prevents models from capturing the semantic vagueness that the individual ratings represent. In this work, we introduce USTS, the first Uncertainty-aware STS dataset with ~15,000 Chinese sentence pairs and 150,000 labels, to study collective human opinions in STS. Analysis reveals that neither a scalar nor a single Gaussian fits a set of observed judgements adequately. We further show that current STS models cannot capture the variance caused by human disagreement on individual instances, but rather reflect the predictive confidence over the aggregate dataset.

Robert M. Corless, Robert H. C. Moir, Marc Moreno Maza et al.

We consider the problem of symbolic-numeric integration of symbolic functions, focusing on rational functions. Using a hybrid method allows the stable yet efficient computation of symbolic antiderivatives while avoiding issues of ill-conditioning to which numerical methods are susceptible. We propose two alternative methods for exact input that compute the rational part of the integral using Hermite reduction and then compute the transcendental part two different ways using a combination of exact integration and efficient numerical computation of roots. The symbolic computation is done within BPAS, or Basic Polynomial Algebra Subprograms, which is a highly optimized environment for polynomial computation on parallel architectures, while the numerical computation is done using the highly optimized multiprecision rootfinding package MPSolve. We show that both methods are forward and backward stable in a structured sense and away from singularities tolerance proportionality is achieved by adjusting the precision of the rootfinding tasks.

Elena Grigorescu, Brendan Juba, Karl Wimmer et al.

Determinantal Point Processes (DPPs) are a widely used probabilistic model for negatively correlated sets. DPPs have been successfully employed in Machine Learning applications to select a diverse, yet representative subset of data. In these applications, a set of parameters that maximize the likelihood of the data is typically desirable. The algorithms used for this task to date either optimize over a limited family of DPPs, or use local improvement heuristics that do not provide theoretical guarantees of optimality. In his seminal work on DPPs in Machine Learning, Kulesza (2011) conjectured that the problem is NP-complete. The lack of a formal proof prompted Brunel et al. (COLT 2017) to suggest that, in opposition to Kulesza's conjecture, there might exist a polynomial-time algorithm for computing a maximum-likelihood DPP. They also presented some preliminary evidence supporting a conjecture that they suggested might lead to such an algorithm. In this work we prove Kulesza's conjecture. In fact, we prove the following stronger hardness of approximation result: even computing a $\left(1-O(\frac{1}{\log^9{N}})\right)$-approximation to the maximum log-likelihood of a DPP on a ground set of $N$ elements is NP-complete. From a technical perspective, we reduce the problem of approximating the maximum log-likelihood of a DPP to solving a gap instance of a \textsc{$3$-Coloring} problem on a hypergraph. This hypergraph is based on the bounded-degree construction of Bogdanov et al. (FOCS 2002), which we enhance using the strong expanders of Alon and Capalbo (FOCS 2007). We demonstrate that if a rank-$3$ DPP achieves near-optimal log-likelihood, its marginal kernel must encode an almost perfect ``vector-coloring" of the hypergraph. Finally, we show that these continuous vectors can be decoded into a proper $3$-coloring after removing a small fraction of ``noisy" edges.

Chengwei Sun, Jiwei Wei, Yujia Wu et al.

Large pre-trained models (LPMs) have demonstrated exceptional performance in diverse natural language processing and computer vision tasks. However, fully fine-tuning these models poses substantial memory challenges, particularly in resource-constrained environments. Parameter-efficient fine-tuning (PEFT) methods, such as LoRA, mitigate this issue by adjusting only a small subset of parameters. Nevertheless, these methods typically employ random initialization for low-rank matrices, which can lead to inefficiencies in gradient descent and diminished generalizability due to suboptimal starting points. To address these limitations, we propose SVFit, a novel PEFT approach that leverages singular value decomposition (SVD) to initialize low-rank matrices using critical singular values as trainable parameters. Specifically, SVFit performs SVD on the pre-trained weight matrix to obtain the best rank-r approximation matrix, emphasizing the most critical singular values that capture over 99% of the matrix's information. These top-r singular values are then used as trainable parameters to scale the fundamental subspaces of the matrix, facilitating rapid domain adaptation. Extensive experiments across various pre-trained models in natural language understanding, text-to-image generation, and image classification tasks reveal that SVFit outperforms LoRA while requiring 16 times fewer trainable parameters.

Zhanglin Wu, Daimeng Wei, Zongyao Li et al.

This paper presents the submission of Huawei Translate Services Center (HW-TSC) to the WMT24 general machine translation (MT) shared task, where we participate in the English to Chinese (en2zh) language pair. Similar to previous years' work, we use training strategies such as regularized dropout, bidirectional training, data diversification, forward translation, back translation, alternated training, curriculum learning, and transductive ensemble learning to train the neural machine translation (NMT) model based on the deep Transformer-big architecture. The difference is that we also use continue pre-training, supervised fine-tuning, and contrastive preference optimization to train the large language model (LLM) based MT model. By using Minimum Bayesian risk (MBR) decoding to select the final translation from multiple hypotheses for NMT and LLM-based MT models, our submission receives competitive results in the final evaluation.

Zhanglin Wu, Yuanchang Luo, Daimeng Wei et al.

This paper presents the submission of Huawei Translation Services Center (HW-TSC) to machine translation tasks of the 20th China Conference on Machine Translation (CCMT 2024). We participate in the bilingual machine translation task and multi-domain machine translation task. For these two translation tasks, we use training strategies such as regularized dropout, bidirectional training, data diversification, forward translation, back translation, alternated training, curriculum learning, and transductive ensemble learning to train neural machine translation (NMT) models based on the deep Transformer-big architecture. Furthermore, to explore whether large language model (LLM) can help improve the translation quality of NMT systems, we use supervised fine-tuning to train llama2-13b as an Automatic post-editing (APE) model to improve the translation results of the NMT model on the multi-domain machine translation task. By using these plyometric strategies, our submission achieves a competitive result in the final evaluation.

Junxin Lu, Tengfei Song, Zhanglin Wu et al.

Text Image Machine Translation (TIMT) aims to translate text embedded in images in the source-language into target-language, requiring synergistic integration of visual perception and linguistic understanding. Existing TIMT methods, whether cascaded pipelines or end-to-end multimodal large language models (MLLMs),struggle with high-resolution text-rich images due to cluttered layouts, diverse fonts, and non-textual distractions, resulting in text omission, semantic drift, and contextual inconsistency. To address these challenges, we propose GLoTran, a global-local dual visual perception framework for MLLM-based TIMT. GLoTran integrates a low-resolution global image with multi-scale region-level text image slices under an instruction-guided alignment strategy, conditioning MLLMs to maintain scene-level contextual consistency while faithfully capturing fine-grained textual details. Moreover, to realize this dual-perception paradigm, we construct GLoD, a large-scale text-rich TIMT dataset comprising 510K high-resolution global-local image-text pairs covering diverse real-world scenarios. Extensive experiments demonstrate that GLoTran substantially improves translation completeness and accuracy over state-of-the-art MLLMs, offering a new paradigm for fine-grained TIMT under high-resolution and text-rich conditions.

Lei Yao, Boxiang He, Shilian Wang et al.

Tag-based physical layer authentication (PLA) has garnered significant attention due to its low complexity and enhanced security. However, existing PLA schemes encounter two challenges. First, unintended user interference, which overlaps with the authentication signal, corrupts the tag and degrades authentication performance. Second, the vulnerability introduced by direct embedding of the raw tag exposes the tag to the adversary and degrades the security. To address these challenges, this paper proposes a novel frozen-tag-based PLA framework. Different from typical schemes that directly embed the uncoded tag into the signal, a well-designed frozen tag is inserted for authentication, where the frozen tag is generated based on the concept of polar codes with the anchor information as information bits and raw tags as frozen bits. Accordingly, the proposed PLA framework offers two principal advantages. First, the authentication performance is improved since the legitimate receiver can decode the frozen tag and mitigate unintended user interference. Second, the authentication process becomes indecipherable to the illegitimate receiver due to the concealment of the raw tags. Furthermore, we conduct a comprehensive analysis of the proposed framework in terms of robustness, security, and compatibility. Theoretical analysis and simulation demonstrate that the proposed frozen-tag-based PLA framework not only enhances the detection performance but also significantly degrades Eve's capability to estimate the raw tags.

Zhanglin Wu, Tengfei Song, Ning Xie et al.

The rapid advancement of large vision-language models (LVLMs) has significantly propelled applications in document understanding, particularly in optical character recognition (OCR) and multilingual translation. However, current evaluations of LVLMs, like the widely used OCRBench, mainly focus on verifying the correctness of their short-text responses and long-text responses with simple layout, while the evaluation of their ability to understand long texts with complex layout design is highly significant but largely overlooked. In this paper, we propose Menu OCR and Translation Benchmark (MOTBench), a specialized evaluation framework emphasizing the pivotal role of menu translation in cross-cultural communication. MOTBench requires LVLMs to accurately recognize and translate each dish, along with its price and unit items on a menu, providing a comprehensive assessment of their visual understanding and language processing capabilities. Our benchmark is comprised of a collection of Chinese and English menus, characterized by intricate layouts, a variety of fonts, and culturally specific elements across different languages, along with precise human annotations. Experiments show that our automatic evaluation results are highly consistent with professional human evaluation. We evaluate a range of publicly available state-of-the-art LVLMs, and through analyzing their output to identify the strengths and weaknesses in their performance, offering valuable insights to guide future advancements in LVLM development. MOTBench is available at https://github.com/gitwzl/MOTBench.

Zhanglin Wu, Tengfei Song, Ning Xie et al.

This paper presents the technical solution proposed by Huawei Translation Service Center (HW-TSC) for the "End-to-End Document Image Machine Translation for Complex Layouts" competition at the 19th International Conference on Document Analysis and Recognition (DIMT25@ICDAR2025). Leveraging state-of-the-art open-source large vision-language model (LVLM), we introduce a training framework that combines multi-task learning with perceptual chain-of-thought to develop a comprehensive end-to-end document translation system. During the inference phase, we apply minimum Bayesian decoding and post-processing strategies to further enhance the system's translation capabilities. Our solution uniquely addresses both OCR-based and OCR-free document image translation tasks within a unified framework. This paper systematically details the training methods, inference strategies, LVLM base models, training data, experimental setups, and results, demonstrating an effective approach to document image machine translation.

Ning Xie, Farley Lai, Derek Doran et al.

We introduce a new inference task - Visual Entailment (VE) - which differs from traditional Textual Entailment (TE) tasks whereby a premise is defined by an image, rather than a natural language sentence as in TE tasks. A novel dataset SNLI-VE (publicly available at https://github.com/necla-ml/SNLI-VE) is proposed for VE tasks based on the Stanford Natural Language Inference corpus and Flickr30k. We introduce a differentiable architecture called the Explainable Visual Entailment model (EVE) to tackle the VE problem. EVE and several other state-of-the-art visual question answering (VQA) based models are evaluated on the SNLI-VE dataset, facilitating grounded language understanding and providing insights on how modern VQA based models perform.

Xudong Wang, Chaoning Zhang, Jiaquan Zhang et al.

Large Language Model (LLM)-driven Multi-Agent Systems (MAS) have demonstrated strong capability in complex reasoning and tool use, and heterogeneous agent pools further broaden the quality--cost trade-off space. Despite these advances, real-world deployment is often constrained by high inference cost, latency, and limited transparency, which hinders scalable and efficient routing. Existing routing strategies typically rely on expensive LLM-based selectors or static policies, and offer limited controllability for semantic-aware routing under dynamic loads and mixed intents, often resulting in unstable performance and inefficient resource utilization. To address these limitations, we propose AMRO-S, an efficient and interpretable routing framework for Multi-Agent Systems (MAS). AMRO-S models MAS routing as a semantic-conditioned path selection problem, enhancing routing performance through three key mechanisms: First, it leverages a supervised fine-tuned (SFT) small language model for intent inference, providing a low-overhead semantic interface for each query; second, it decomposes routing memory into task-specific pheromone specialists, reducing cross-task interference and optimizing path selection under mixed workloads; finally, it employs a quality-gated asynchronous update mechanism to decouple inference from learning, optimizing routing without increasing latency. Extensive experiments on five public benchmarks and high-concurrency stress tests demonstrate that AMRO-S consistently improves the quality--cost trade-off over strong routing baselines, while providing traceable routing evidence through structured pheromone patterns.

Jingya Wang, Xin Deng, Wenjie Wei et al.

Leveraging the event-driven paradigm, Spiking Neural Networks (SNNs) offer a promising approach for constructing energy-efficient Transformer architectures. Compared to directly trained Spiking Transformers, ANN-to-SNN conversion methods bypass the high training costs. However, existing methods still suffer from notable limitations, failing to effectively handle nonlinear operations in Transformer architectures and requiring additional fine-tuning processes for pre-trained ANNs. To address these issues, we propose a high-performance and training-free ANN-to-SNN conversion framework tailored for Transformer architectures. Specifically, we introduce a Multi-basis Exponential Decay (MBE) neuron, which employs an exponential decay strategy and multi-basis encoding method to efficiently approximate various nonlinear operations. It removes the requirement for weight modifications in pre-trained ANNs. Extensive experiments across diverse tasks (CV, NLU, NLG) and mainstream Transformer architectures (ViT, RoBERTa, GPT-2) demonstrate that our method achieves near-lossless conversion accuracy with significantly lower latency. This provides a promising pathway for the efficient and scalable deployment of Spiking Transformers in real-world applications.

Qian Liang, Yujia Wu, Kuncheng Li et al.

Multimodal Large Language Models (MLLMs) with unified architectures excel across a wide range of vision-language tasks, yet aligning them with personalized image generation remains a significant challenge. Existing methods for MLLMs are frequently subject-specific, demanding a data-intensive fine-tuning process for every new subject, which limits their scalability. In this paper, we introduce MM-R1, a framework that integrates a cross-modal Chain-of-Thought (X-CoT) reasoning strategy to unlock the inherent potential of unified MLLMs for personalized image generation. Specifically, we structure personalization as an integrated visual reasoning and generation process: (1) grounding subject concepts by interpreting and understanding user-provided images and contextual cues, and (2) generating personalized images conditioned on both the extracted subject representations and user prompts. To further enhance the reasoning capability, we adopt Grouped Reward Proximal Policy Optimization (GRPO) to explicitly align the generation. Experiments demonstrate that MM-R1 unleashes the personalization capability of unified MLLMs to generate images with high subject fidelity and strong text alignment in a zero-shot manner.

Shuchang Zhou, Jiwei Wei, Shiyuan He et al.

Prompt tuning has become a popular strategy for adapting Vision-Language Models (VLMs) to zero/few-shot visual recognition tasks. Some prompting techniques introduce prior knowledge due to its richness, but when learnable tokens are randomly initialized and disconnected from prior knowledge, they tend to overfit on seen classes and struggle with domain shifts for unseen ones. To address this issue, we propose the InPK model, which infuses class-specific prior knowledge into the learnable tokens during initialization, thus enabling the model to explicitly focus on class-relevant information. Furthermore, to mitigate the weakening of class information by multi-layer encoders, we continuously reinforce the interaction between learnable tokens and prior knowledge across multiple feature levels. This progressive interaction allows the learnable tokens to better capture the fine-grained differences and universal visual concepts within prior knowledge, enabling the model to extract more discriminative and generalized text features. Even for unseen classes, the learned interaction allows the model to capture their common representations and infer their appropriate positions within the existing semantic structure. Moreover, we introduce a learnable text-to-vision projection layer to accommodate the text adjustments, ensuring better alignment of visual-text semantics. Extensive experiments on 11 recognition datasets show that InPK significantly outperforms state-of-the-art methods in multiple zero/few-shot image classification tasks.

Jianing Zhou, Han Li, Shuai Zhang et al. · amazon-science

Handling lengthy context is crucial for enhancing the recognition and understanding capabilities of multimodal large language models (MLLMs) in applications such as processing high-resolution images or high frame rate videos. The rise in image resolution and frame rate substantially increases computational demands due to the increased number of input tokens. This challenge is further exacerbated by the quadratic complexity with respect to sequence length of the self-attention mechanism. Most prior works either pre-train models with long contexts, overlooking the efficiency problem, or attempt to reduce the context length via downsampling (e.g., identify the key image patches or frames) to decrease the context length, which may result in information loss. To circumvent this issue while keeping the remarkable effectiveness of MLLMs, we propose a novel approach using a hybrid transformer-MAMBA model to efficiently handle long contexts in multimodal applications. Our multimodal model can effectively process long context input exceeding 100k tokens, outperforming existing models across various benchmarks. Remarkably, our model enhances inference efficiency for high-resolution images and high-frame-rate videos by about 4 times compared to current models, with efficiency gains increasing as image resolution or video frames rise. Furthermore, our model is the first to be trained on low-resolution images or low-frame-rate videos while being capable of inference on high-resolution images and high-frame-rate videos, offering flexibility for inference in diverse scenarios.

Mustafa Ocal, Ning Xie, Mark Finlayson

A timeline provides a total ordering of events and times, and is useful for a number of natural language understanding tasks. However, qualitative temporal graphs that can be derived directly from text -- such as TimeML annotations -- usually explicitly reveal only partial orderings of events and times. In this work, we apply prior work on solving point algebra problems to the task of extracting timelines from TimeML annotated texts, and develop an exact, end-to-end solution which we call TLEX (TimeLine EXtraction). TLEX transforms TimeML annotations into a collection of timelines arranged in a trunk-and-branch structure. Like what has been done in prior work, TLEX checks the consistency of the temporal graph and solves it; however, it adds two novel functionalities. First, it identifies specific relations involved in an inconsistency (which could then be manually corrected) and, second, TLEX performs a novel identification of sections of the timelines that have indeterminate order, information critical for downstream tasks such as aligning events from different timelines. We provide detailed descriptions and analysis of the algorithmic components in TLEX, and conduct experimental evaluations by applying TLEX to 385 TimeML annotated texts from four corpora. We show that 123 of the texts are inconsistent, 181 of them have more than one ``real world'' or main timeline, and there are 2,541 indeterminate sections across all four corpora. A sampling evaluation showed that TLEX is 98--100% accurate with 95% confidence along five dimensions: the ordering of time-points, the number of main timelines, the placement of time-points on main versus subordinate timelines, the connecting point of branch timelines, and the location of the indeterminate sections. We provide a reference implementation of TLEX, the extracted timelines for all texts, and the manual corrections of the inconsistent texts.

Zhitao Liu, Zengyu Liu, Jiwei Wei et al.

3D cross-modal retrieval is gaining attention in the multimedia community. Central to this topic is learning a joint embedding space to represent data from different modalities, such as images, 3D point clouds, and polygon meshes, to extract modality-invariant and discriminative features. Hence, the performance of cross-modal retrieval methods heavily depends on the representational capacity of this embedding space. Existing methods treat all instances equally, applying the same penalty strength to instances with varying degrees of difficulty, ignoring the differences between instances. This can result in ambiguous convergence or local optima, severely compromising the separability of the feature space. To address this limitation, we propose an Instance-Variant loss to assign different penalty strengths to different instances, improving the space separability. Specifically, we assign different penalty weights to instances positively related to their intra-class distance. Simultaneously, we reduce the cross-modal discrepancy between features by learning a shared weight vector for the same class data from different modalities. By leveraging the Gaussian RBF kernel to evaluate sample similarity, we further propose an Intra-Class loss function that minimizes the intra-class distance among same-class instances. Extensive experiments on three 3D cross-modal datasets show that our proposed method surpasses recent state-of-the-art approaches.

Zhitao Liu, Ning Xie, Guobiao Yang et al.

Quantitative analysis of Game User eXperience (GUX) is important to the game industry. Different from the typical questionnaire analysis, this paper focuses on the computational analysis of GUX. We aim to analyze the relationship between game and players using the multi-modal data including physiological data and game process data. We theoretically extend the Flow model from the classic skill-and-challenge plane by expanding new dimension on motivation, which is the result of the multi-modal data analysis on affect, and physiological data. We call this 3D Flow as Motivational Flow, MovFlow. Meanwhile, we implement a quantitative GUX Analysis System (GUXAS), which can predict the player's in-game experience state by only using game process data. It analyzes the correlation among not only in-game state, but the player's psychological-and-physiological reaction in the entire interactive game-play process. The experiments demonstrated our MovFlow model efficiently distinguished the users' in-game experience states from the perspective of GUX.

Zhitao Liu, Jinke Shi, Junhao He et al.

To adapt to different environments, human circadian rhythms will be constantly adjusted as the environment changes, which follows the principle of survival of the fittest. According to this principle, objective factors (such as circadian rhythms, and light intensity) can be utilized to control time perception. The subjective judgment on the estimation of elapsed time is called time perception. In the physical world, factors that can affect time perception, represented by illumination, are called the Zeitgebers. In recent years, with the development of Virtual Reality (VR) technology, effective control of zeitgebers has become possible, which is difficult to achieve in the physical world. Based on previous studies, this paper deeply explores the actual performance in VR environment of four types of time zeitgebers (music, color, cognitive load, and concentration) that have been proven to have a certain impact on time perception in the physical world. It discusses the study of the measurement of the difference between human time perception and objective escaped time in the physical world.

Huidong Liu, Shaoyuan Xu, Jinmiao Fu et al.

Modern Web systems such as social media and e-commerce contain rich contents expressed in images and text. Leveraging information from multi-modalities can improve the performance of machine learning tasks such as classification and recommendation. In this paper, we propose the Cross-Modality Attention Contrastive Language-Image Pre-training (CMA-CLIP), a new framework which unifies two types of cross-modality attentions, sequence-wise attention and modality-wise attention, to effectively fuse information from image and text pairs. The sequence-wise attention enables the framework to capture the fine-grained relationship between image patches and text tokens, while the modality-wise attention weighs each modality by its relevance to the downstream tasks. In addition, by adding task specific modality-wise attentions and multilayer perceptrons, our proposed framework is capable of performing multi-task classification with multi-modalities. We conduct experiments on a Major Retail Website Product Attribute (MRWPA) dataset and two public datasets, Food101 and Fashion-Gen. The results show that CMA-CLIP outperforms the pre-trained and fine-tuned CLIP by an average of 11.9% in recall at the same level of precision on the MRWPA dataset for multi-task classification. It also surpasses the state-of-the-art method on Fashion-Gen Dataset by 5.5% in accuracy and achieves competitive performance on Food101 Dataset. Through detailed ablation studies, we further demonstrate the effectiveness of both cross-modality attention modules and our method's robustness against noise in image and text inputs, which is a common challenge in practice.

Junhao Zeng, Qingchun Zhang, Ning Xie et al.

The development of intelligent tutoring system has greatly influenced the way students learn and practice, which increases their learning efficiency. The intelligent tutoring system must model learners' mastery of the knowledge before providing feedback and advices to learners, so one class of algorithm called "knowledge tracing" is surely important. This paper proposed Deep Self-Attentive Knowledge Tracing (DSAKT) based on the data of PTA, an online assessment system used by students in many universities in China, to help these students learn more efficiently. Experimentation on the data of PTA shows that DSAKT outperforms the other models for knowledge tracing an improvement of AUC by 2.1% on average, and this model also has a good performance on the ASSIST dataset.

Ao Luo, Ning Xie, Zhijia Tao et al.

Human motion is a key function to communicate information. In the application, short-form mobile video is so popular all over the world such as Tik Tok. The users would like to add more VFX so as to pursue creativity and personlity. Many special effects are added on the short video platform. These gives the users more possibility to show off these personality. The common and traditional way is to create the template of VFX. However, in order to synthesis the perfect, the users have to tedious attempt to grasp the timing and rhythm of new templates. It is not easy-to-use especially for the mobile app. This paper aims to change the VFX synthesis by motion driven instead of the traditional template matching. We propose the AI method to improve this VFX synthesis. In detail, in order to add the special effect on the human body. The skeleton extraction is essential in this system. We also propose a novel form of LSTM to find out the user's intention by action recognition. The experiment shows that our system enables to generate VFX for short video more easier and efficient.

Mahdi Cheraghchi, Elena Grigorescu, Brendan Juba et al.

We introduce and study the model of list learning with attribute noise. Learning with attribute noise was introduced by Shackelford and Volper (COLT 1988) as a variant of PAC learning, in which the algorithm has access to noisy examples and uncorrupted labels, and the goal is to recover an accurate hypothesis. Sloan (COLT 1988) and Goldman and Sloan (Algorithmica 1995) discovered information-theoretic limits to learning in this model, which have impeded further progress. In this article we extend the model to that of list learning, drawing inspiration from the list-decoding model in coding theory, and its recent variant studied in the context of learning. On the positive side, we show that sparse conjunctions can be efficiently list learned under some assumptions on the underlying ground-truth distribution. On the negative side, our results show that even in the list-learning model, efficient learning of parities and majorities is not possible regardless of the representation used.

Gabrielle Ras, Ning Xie, Marcel van Gerven et al.

Deep neural networks (DNNs) have become a proven and indispensable machine learning tool. As a black-box model, it remains difficult to diagnose what aspects of the model's input drive the decisions of a DNN. In countless real-world domains, from legislation and law enforcement to healthcare, such diagnosis is essential to ensure that DNN decisions are driven by aspects appropriate in the context of its use. The development of methods and studies enabling the explanation of a DNN's decisions has thus blossomed into an active, broad area of research. A practitioner wanting to study explainable deep learning may be intimidated by the plethora of orthogonal directions the field has taken. This complexity is further exacerbated by competing definitions of what it means ``to explain'' the actions of a DNN and to evaluate an approach's ``ability to explain''. This article offers a field guide to explore the space of explainable deep learning aimed at those uninitiated in the field. The field guide: i) Introduces three simple dimensions defining the space of foundational methods that contribute to explainable deep learning, ii) discusses the evaluations for model explanations, iii) places explainability in the context of other related deep learning research areas, and iv) finally elaborates on user-oriented explanation designing and potential future directions on explainable deep learning. We hope the guide is used as an easy-to-digest starting point for those just embarking on research in this field.

Yunlian Lv, Ning Xie, Yimin Shi et al.

Embodied artificial intelligence (AI) tasks shift from tasks focusing on internet images to active settings involving embodied agents that perceive and act within 3D environments. In this paper, we investigate the target-driven visual navigation using deep reinforcement learning (DRL) in 3D indoor scenes, whose navigation task aims to train an agent that can intelligently make a series of decisions to arrive at a pre-specified target location from any possible starting positions only based on egocentric views. However, most navigation methods currently struggle against several challenging problems, such as data efficiency, automatic obstacle avoidance, and generalization. Generalization problem means that agent does not have the ability to transfer navigation skills learned from previous experience to unseen targets and scenes. To address these issues, we incorporate two designs into classic DRL framework: attention on 3D knowledge graph (KG) and target skill extension (TSE) module. On the one hand, our proposed method combines visual features and 3D spatial representations to learn navigation policy. On the other hand, TSE module is used to generate sub-targets which allow agent to learn from failures. Specifically, our 3D spatial relationships are encoded through recently popular graph convolutional network (GCN). Considering the real world settings, our work also considers open action and adds actionable targets into conventional navigation situations. Those more difficult settings are applied to test whether DRL agent really understand its task, navigating environment, and can carry out reasoning. Our experiments, performed in the AI2-THOR, show that our model outperforms the baselines in both SR and SPL metrics, and improves generalization ability across targets and scenes.

Ning Xie, Ji Hu, Junjie Chen et al.

This paper concerns the problem of detecting the use of information hiding at anti-copying 2D barcodes. Prior hidden information detection schemes are either heuristicbased or Machine Learning (ML) based. The key limitation of prior heuristics-based schemes is that they do not answer the fundamental question of why the information hidden at a 2D barcode can be detected. The key limitation of prior MLbased information schemes is that they lack robustness because a printed 2D barcode is very much environmentally dependent, and thus an information hiding detection scheme trained in one environment often does not work well in another environment. In this paper, we propose two hidden information detection schemes at the existing anti-copying 2D barcodes. The first scheme is to directly use the pixel distance to detect the use of an information hiding scheme in a 2D barcode, referred as to the Pixel Distance Based Detection (PDBD) scheme. The second scheme is first to calculate the variance of the raw signal and the covariance between the recovered signal and the raw signal, and then based on the variance results, detects the use of information hiding scheme in a 2D barcode, referred as to the Pixel Variance Based Detection (PVBD) scheme. Moreover, we design advanced IC attacks to evaluate the security of two existing anti-copying 2D barcodes. We implemented our schemes and conducted extensive performance comparison between our schemes and prior schemes under different capturing devices, such as a scanner and a camera phone. Our experimental results show that the PVBD scheme can correctly detect the existence of the hidden information at both the 2LQR code and the LCAC 2D barcode. Moreover, the probability of successfully attacking of our IC attacks achieves 0.6538 for the 2LQR code and 1 for the LCAC 2D barcode.

Ning Xie, Qiqi Zhang, Ji Hu et al.

In this paper, for overcoming the drawbacks of the prior approaches, such as low generality, high cost, and high overhead, we propose a Low-Cost Anti-Copying (LCAC) 2D barcode by exploiting the difference between the noise characteristics of legal and illegal channels. An embedding strategy is proposed, and for a variant of it, we also make the corresponding analysis. For accurately evaluating the performance of our approach, a theoretical model of the noise in an illegal channel is established by using a generalized Gaussian distribution. By comparing with the experimental results based on various printers, scanners, and a mobile phone, it can be found that the sample histogram and curve fitting of the theoretical model match well, so it can be concluded that the theoretical model works well. For evaluating the security of the proposed LCAC code, besides the direct-copying (DC) attack, the improved version, which is the synthesized-copying (SC) attack, is also considered in this paper. Based on the theoretical model, we build a prediction function to optimize the parameters of our approach. The parameters optimization incorporates the covertness requirement, the robustness requirement and a tradeoff between the production cost and the cost of illegally-copying attacks together. The experimental results show that the proposed LCAC code with two printers and two scanners can detect the DC attack effectively and resist the SC attack up to the access of 14 legal copies.

Farley Lai, Ning Xie, Derek Doran et al.

In this paper, we introduce a contextual grounding approach that captures the context in corresponding text entities and image regions to improve the grounding accuracy. Specifically, the proposed architecture accepts pre-trained text token embeddings and image object features from an off-the-shelf object detector as input. Additional encoding to capture the positional and spatial information can be added to enhance the feature quality. There are separate text and image branches facilitating respective architectural refinements for different modalities. The text branch is pre-trained on a large-scale masked language modeling task while the image branch is trained from scratch. Next, the model learns the contextual representations of the text tokens and image objects through layers of high-order interaction respectively. The final grounding head ranks the correspondence between the textual and visual representations through cross-modal interaction. In the evaluation, we show that our model achieves the state-of-the-art grounding accuracy of 71.36% over the Flickr30K Entities dataset. No additional pre-training is necessary to deliver competitive results compared with related work that often requires task-agnostic and task-specific pre-training on cross-modal dadasets. The implementation is publicly available at https://gitlab.com/necla-ml/grounding.

Ning Xie, Farley Lai, Derek Doran et al.

Existing visual reasoning datasets such as Visual Question Answering (VQA), often suffer from biases conditioned on the question, image or answer distributions. The recently proposed CLEVR dataset addresses these limitations and requires fine-grained reasoning but the dataset is synthetic and consists of similar objects and sentence structures across the dataset. In this paper, we introduce a new inference task, Visual Entailment (VE) - consisting of image-sentence pairs whereby a premise is defined by an image, rather than a natural language sentence as in traditional Textual Entailment tasks. The goal of a trained VE model is to predict whether the image semantically entails the text. To realize this task, we build a dataset SNLI-VE based on the Stanford Natural Language Inference corpus and Flickr30k dataset. We evaluate various existing VQA baselines and build a model called Explainable Visual Entailment (EVE) system to address the VE task. EVE achieves up to 71% accuracy and outperforms several other state-of-the-art VQA based models. Finally, we demonstrate the explainability of EVE through cross-modal attention visualizations. The SNLI-VE dataset is publicly available at https://github.com/ necla-ml/SNLI-VE.

Monireh Ebrahimi, Md Kamruzzaman Sarker, Federico Bianchi et al.

Semantic Web knowledge representation standards, and in particular RDF and OWL, often come endowed with a formal semantics which is considered to be of fundamental importance for the field. Reasoning, i.e., the drawing of logical inferences from knowledge expressed in such standards, is traditionally based on logical deductive methods and algorithms which can be proven to be sound and complete and terminating, i.e. correct in a very strong sense. For various reasons, though, in particular, the scalability issues arising from the ever-increasing amounts of Semantic Web data available and the inability of deductive algorithms to deal with noise in the data, it has been argued that alternative means of reasoning should be investigated which bear high promise for high scalability and better robustness. From this perspective, deductive algorithms can be considered the gold standard regarding correctness against which alternative methods need to be tested. In this paper, we show that it is possible to train a Deep Learning system on RDF knowledge graphs, such that it is able to perform reasoning over new RDF knowledge graphs, with high precision and recall compared to the deductive gold standard.

Ning Xie, Md Kamruzzaman Sarker, Derek Doran et al.

Many current methods to interpret convolutional neural networks (CNNs) use visualization techniques and words to highlight concepts of the input seemingly relevant to a CNN's decision. The methods hypothesize that the recognition of these concepts are instrumental in the decision a CNN reaches, but the nature of this relationship has not been well explored. To address this gap, this paper examines the quality of a concept's recognition by a CNN and the degree to which the recognitions are associated with CNN decisions. The study considers a CNN trained for scene recognition over the ADE20k dataset. It uses a novel approach to find and score the strength of minimally distributed representations of input concepts (defined by objects in scene images) across late stage feature maps. Subsequent analysis finds evidence that concept recognition impacts decision making. Strong recognition of concepts frequently-occurring in few scenes are indicative of correct decisions, but recognizing concepts common to many scenes may mislead the network.

Md Kamruzzaman Sarker, Ning Xie, Derek Doran et al.

The ever increasing prevalence of publicly available structured data on the World Wide Web enables new applications in a variety of domains. In this paper, we provide a conceptual approach that leverages such data in order to explain the input-output behavior of trained artificial neural networks. We apply existing Semantic Web technologies in order to provide an experimental proof of concept.

Voot Tangkaratt, Ning Xie, Masashi Sugiyama

Regression aims at estimating the conditional mean of output given input. However, regression is not informative enough if the conditional density is multimodal, heteroscedastic, and asymmetric. In such a case, estimating the conditional density itself is preferable, but conditional density estimation (CDE) is challenging in high-dimensional space. A naive approach to coping with high-dimensionality is to first perform dimensionality reduction (DR) and then execute CDE. However, such a two-step process does not perform well in practice because the error incurred in the first DR step can be magnified in the second CDE step. In this paper, we propose a novel single-shot procedure that performs CDE and DR simultaneously in an integrated way. Our key idea is to formulate DR as the problem of minimizing a squared-loss variant of conditional entropy, and this is solved via CDE. Thus, an additional CDE step is not needed after DR. We demonstrate the usefulness of the proposed method through extensive experiments on various datasets including humanoid robot transition and computer art.

Ning Xie, Hirotaka Hachiya, Masashi Sugiyama

Oriental ink painting, called Sumi-e, is one of the most appealing painting styles that has attracted artists around the world. Major challenges in computer-based Sumi-e simulation are to abstract complex scene information and draw smooth and natural brush strokes. To automatically find such strokes, we propose to model the brush as a reinforcement learning agent, and learn desired brush-trajectories by maximizing the sum of rewards in the policy search framework. We also provide elaborate design of actions, states, and rewards tailored for a Sumi-e agent. The effectiveness of our proposed approach is demonstrated through simulated Sumi-e experiments.