arXiv subjects

Jun Xiao

Publications and source records attributed to Jun Xiao.

At least 235 records · Page 13Linked to original sources

arXiv2021

Human-like Controllable Image Captioning with Verb-specific Semantic Roles

Controllable Image Captioning (CIC) -- generating image descriptions following designated control signals -- has received unprecedented attention over the last few years. To emulate the human ability in controlling caption generation, current CIC studies focus exclusively on control signals concerning objective properties, such as contents of interest or descriptive patterns. However, we argue that almost all existing objective control signals have overlooked two indispensable characteristics of an ideal control signal: 1) Event-compatible: all visual contents referred to in a single sentence should be compatible with the described activity. 2) Sample-suitable: the control signals should be suitable for a specific image sample. To this end, we propose a new control signal for CIC: Verb-specific Semantic Roles (VSR). VSR consists of a verb and some semantic roles, which represents a targeted activity and the roles of entities involved in this activity. Given a designated VSR, we first train a grounded semantic role labeling (GSRL) model to identify and ground all entities for each role. Then, we propose a semantic structure planner (SSP) to learn human-like descriptive semantic structures. Lastly, we use a role-shift captioning model to generate the captions. Extensive experiments and ablations demonstrate that our framework can achieve better controllability than several strong baselines on two challenging CIC benchmarks. Besides, we can generate multi-level diverse captions easily. The code is available at: https://github.com/mad-red/VSR-guided-CIC.

cs.CV↗

arXiv2021

Benchmarking calculations with spectroscopic accuracy of level energies and wavelengths in W LVII - W LXII tungsten ions

Atomic properties of $n=3$ states of the W$^{56+}$ $-$ W$^{61+}$ ions are systematically investigated through two state-of-the-art methods, namely, the second-order many-body perturbation theory, and the multi-configuration Dirac-Hartree-Fock method combined with the relativistic configuration interaction approach. The contributions of valence-valence and core-valence electron correlations, the Breit interaction, the higher-order retardation correction beyond the Breit interaction through the transverse photon interaction, and the quantum electrodynamical corrections to the excitation energies are studied in detail. The excitation energies and wavelengths obtained with the two methods agree with each other within \approx 0.01 %. The present results achieve spectroscopic accuracy and provide a benchmark test for various applications and other theoretical calculations of W$^{56+}$ $-$ W$^{61+}$ ions. They will assist spectroscopists in their assignment and direct identification of observed lines in complex spectra.

physics.atom-ph↗

arXiv2021

Ref-NMS: Breaking Proposal Bottlenecks in Two-Stage Referring Expression Grounding

The prevailing framework for solving referring expression grounding is based on a two-stage process: 1) detecting proposals with an object detector and 2) grounding the referent to one of the proposals. Existing two-stage solutions mostly focus on the grounding step, which aims to align the expressions with the proposals. In this paper, we argue that these methods overlook an obvious mismatch between the roles of proposals in the two stages: they generate proposals solely based on the detection confidence (i.e., expression-agnostic), hoping that the proposals contain all right instances in the expression (i.e., expression-aware). Due to this mismatch, current two-stage methods suffer from a severe performance drop between detected and ground-truth proposals. To this end, we propose Ref-NMS, which is the first method to yield expression-aware proposals at the first stage. Ref-NMS regards all nouns in the expression as critical objects, and introduces a lightweight module to predict a score for aligning each box with a critical object. These scores can guide the NMS operation to filter out the boxes irrelevant to the expression, increasing the recall of critical objects, resulting in a significantly improved grounding performance. Since Ref- NMS is agnostic to the grounding step, it can be easily integrated into any state-of-the-art two-stage method. Extensive ablation studies on several backbones, benchmarks, and tasks consistently demonstrate the superiority of Ref-NMS. Codes are available at: https://github.com/ChopinSharp/ref-nms.

cs.CV↗

arXiv2021

Kinetic Energy Distribution of Fragments for Thermal Neutron-Induced $^{235}$U and $^{239}$Pu Fission Reactions

Focused on the generation and evolution of vast complementary pairs of the primary fission fragments at scission moment, Dinuclear and Statistical Model (DSM) is proposed. (1) It is assumed that the fissile nucleus elongates along a symmetric coaxis until it breaks into two primary fission fragments. (2) Every complementary pair of the primary fission fragments is approximatively described as two ellipsoids with large deformation at scission moment. (3) The kinetic energy in every complementary pair of the primary fragments is mainly provided by Coulomb repulsion, which is explicitly expressed through strict six-dimensional integrals. (4) Only three phenomenological coefficients are obtained to globally describe the quadrupole deformation parameters of arbitrary primary fragments both for $^{235}$U($n_{th}, f$) and $^{239}$Pu($n_{th}, f$) reactions, on the basis of the common characteristics of the measured data, such as mass and charge distributions, kinetic energy distributions. In the framework of DSM, the explicit average total kinetic energy distribution $\overline{TKE}(A)$ and the average kinetic energy distribution $\overline{KE}(A)$ are consistently represented. The theoretical results in this paper agree well with the experimental data. Furthermore, this model is expected as the reliable approach to generally evaluate the corresponding observebles for thermal neutron-induced fission of actinides.

nucl-th↗

arXiv2021

Probing Multiple Electric Dipole Forbidden Optical Transitions in Highly Charged Nickel Ions

Highly charged ions (HCIs) are promising candidates for the next generation of atomic clocks, owing to their tightly bound electron cloud, which significantly suppresses the common environmental disturbances to the quantum oscillator. Here we propose and pursue an experimental strategy that, while focusing on various HCIs of a single atomic element, keeps the number of candidate clock transitions as large as possible. Following this strategy, we identify four adjacent charge states of nickel HCIs that offer as many as six optical transitions. Experimentally, we demonstrated the essential capability of producing these ions in the low-energy compact Shanghai-Wuhan Electron Beam Ion Trap. We measured the wavelengths of four magnetic-dipole ($M$1) and one electric-quadrupole ($E$2) clock transitions with an accuracy of several ppm with a novel calibration method; two of these lines were observed and characterized for the first time in controlled laboratory settings. Compared to the earlier determinations, our measurements improved wavelength accuracy by an order of magnitude. Such measurements are crucial for constraining the range of laser wavelengths for finding the "needle in a haystack" narrow lines. In addition, we calculated frequencies and quality factors, evaluated sensitivity of these six transitions to the hypothetical variation of the electromagnetic fine structure constant $α$ needed for fundamental physics applications. We argue that all the six transitions in nickel HCIs offer intrinsic immunity to all common perturbations of quantum oscillators, and one of them has the projected fractional frequency uncertainty down to the remarkable level of 10$^{-19}$.

physics.atom-ph↗

arXiv2020

ROBY: Evaluating the Robustness of a Deep Model by its Decision Boundaries

With the successful application of deep learning models in many real-world tasks, the model robustness becomes more and more critical. Often, we evaluate the robustness of the deep models by attacking them with purposely generated adversarial samples, which is computationally costly and dependent on the specific attackers and the model types. This work proposes a generic evaluation metric ROBY, a novel attack-independent robustness measure based on the model's decision boundaries. Independent of adversarial samples, ROBY uses the inter-class and intra-class statistic features to capture the features of the model's decision boundaries. We experimented on ten state-of-the-art deep models and showed that ROBY matches the robustness gold standard of attack success rate (ASR) by a strong first-order generic attacker. with only 1% of time cost. To the best of our knowledge, ROBY is the first lightweight attack-independent robustness evaluation metric that can be applied to a wide range of deep models. The code of ROBY is open sourced at https://github.com/baaaad/ROBY-Evaluating-the-Robustness-of-a-Deep-Model-by-its-Decision-Boundaries.

cs.LG↗

arXiv2020

Federated Unsupervised Representation Learning

To leverage enormous unlabeled data on distributed edge devices, we formulate a new problem in federated learning called Federated Unsupervised Representation Learning (FURL) to learn a common representation model without supervision while preserving data privacy. FURL poses two new challenges: (1) data distribution shift (Non-IID distribution) among clients would make local models focus on different categories, leading to the inconsistency of representation spaces. (2) without the unified information among clients in FURL, the representations across clients would be misaligned. To address these challenges, we propose Federated Constrastive Averaging with dictionary and alignment (FedCA) algorithm. FedCA is composed of two key modules: (1) dictionary module to aggregate the representations of samples from each client and share with all clients for consistency of representation space and (2) alignment module to align the representation of each client on a base model trained on a public data. We adopt the contrastive loss for local model training. Through extensive experiments with three evaluation protocols in IID and Non-IID settings, we demonstrate that FedCA outperforms all baselines with significant margins.

cs.LG↗

arXiv2020

Topic Adaptation and Prototype Encoding for Few-Shot Visual Storytelling

Visual Storytelling~(VIST) is a task to tell a narrative story about a certain topic according to the given photo stream. The existing studies focus on designing complex models, which rely on a huge amount of human-annotated data. However, the annotation of VIST is extremely costly and many topics cannot be covered in the training dataset due to the long-tail topic distribution. In this paper, we focus on enhancing the generalization ability of the VIST model by considering the few-shot setting. Inspired by the way humans tell a story, we propose a topic adaptive storyteller to model the ability of inter-topic generalization. In practice, we apply the gradient-based meta-learning algorithm on multi-modal seq2seq models to endow the model the ability to adapt quickly from topic to topic. Besides, We further propose a prototype encoding structure to model the ability of intra-topic derivation. Specifically, we encode and restore the few training story text to serve as a reference to guide the generation at inference time. Experimental results show that topic adaptation and prototype encoding structure mutually bring benefit to the few-shot model on BLEU and METEOR metric. The further case study shows that the stories generated after few-shot adaptation are more relative and expressive.

cs.CL↗

arXiv2020

Berry curvature memory through electrically driven stacking transitions

In two-dimensional layered quantum materials, the stacking order of the layers determines both the crystalline symmetry and electronic properties such as the Berry curvature, topology and electron correlation. Electrical stimuli can influence quasiparticle interactions and the free-energy landscape, making it possible to dynamically modify the stacking order and reveal hidden structures that host different quantum properties. Here we demonstrate electrically driven stacking transitions that can be applied to design nonvolatile memory based on Berry curvature in few-layer WTe$_2$. The interplay of out-of-plane electric fields and electrostatic doping controls in-plane interlayer sliding and creates multiple polar and centrosymmetric stacking orders. In situ nonlinear Hall transport reveals such stacking rearrangements result in a layer-parity-selective Berry curvature memory in momentum space, where the sign reversal of the Berry curvature and its dipole only occurs in odd-layer crystals. Our findings open an avenue towards exploring coupling between topology, electron correlations, and ferroelectricity in hidden stacking orders and demonstrate a new low-energy-cost, electrically controlled topological memory in the atomically thin limit.

cond-mat.mtrl-sci↗

arXiv2020

Hierarchical Fashion Graph Network for Personalized Outfit Recommendation

Fashion outfit recommendation has attracted increasing attentions from online shopping services and fashion communities.Distinct from other scenarios (e.g., social networking or content sharing) which recommend a single item (e.g., a friend or picture) to a user, outfit recommendation predicts user preference on a set of well-matched fashion items.Hence, performing high-quality personalized outfit recommendation should satisfy two requirements -- 1) the nice compatibility of fashion items and 2) the consistence with user preference. However, present works focus mainly on one of the requirements and only consider either user-outfit or outfit-item relationships, thereby easily leading to suboptimal representations and limiting the performance. In this work, we unify two tasks, fashion compatibility modeling and personalized outfit recommendation. Towards this end, we develop a new framework, Hierarchical Fashion Graph Network(HFGN), to model relationships among users, items, and outfits simultaneously. In particular, we construct a hierarchical structure upon user-outfit interactions and outfit-item mappings. We then get inspirations from recent graph neural networks, and employ the embedding propagation on such hierarchical graph, so as to aggregate item information into an outfit representation, and then refine a user's representation via his/her historical outfits. Furthermore, we jointly train these two tasks to optimize these representations. To demonstrate the effectiveness of HFGN, we conduct extensive experiments on a benchmark dataset, and HFGN achieves significant improvements over the state-of-the-art compatibility matching models like NGNN and outfit recommenders like FHN.

cs.IR↗

arXiv2020

Strain-Induced Room-Temperature Ferroelectricity in SrTiO$_3$ Membranes

Advances in complex oxide heteroepitaxy have highlighted the enormous potential of utilizing strain engineering via lattice mismatch to control ferroelectricity in thin-film heterostructures. This approach, however, lacks the ability to produce large and continuously variable strain states, thus limiting the potential for designing and tuning the desired properties of ferroelectric films. Here, we observe and explore dynamic strain-induced ferroelectricity in SrTiO$_3$ by laminating freestanding oxide films onto a stretchable polymer substrate. Using a combination of scanning probe microscopy, optical second harmonic generation measurements, and atomistic modeling, we demonstrate robust room-temperature ferroelectricity in SrTiO$_3$ with 2.0% uniaxial tensile strain, corroborated by the notable features of 180° ferroelectric domains and an extrapolated transition temperature of 400 K. Our work reveals the enormous potential of employing oxide membranes to create and enhance ferroelectricity in environmentally benign lead-free oxides, which hold great promise for applications ranging from non-volatile memories and microwave electronics.

cond-mat.mtrl-sci↗

arXiv2020

Counterfactual Samples Synthesizing for Robust Visual Question Answering

Despite Visual Question Answering (VQA) has realized impressive progress over the last few years, today's VQA models tend to capture superficial linguistic correlations in the train set and fail to generalize to the test set with different QA distributions. To reduce the language biases, several recent works introduce an auxiliary question-only model to regularize the training of targeted VQA model, and achieve dominating performance on VQA-CP. However, since the complexity of design, current methods are unable to equip the ensemble-based models with two indispensable characteristics of an ideal VQA model: 1) visual-explainable: the model should rely on the right visual regions when making decisions. 2) question-sensitive: the model should be sensitive to the linguistic variations in question. To this end, we propose a model-agnostic Counterfactual Samples Synthesizing (CSS) training scheme. The CSS generates numerous counterfactual training samples by masking critical objects in images or words in questions, and assigning different ground-truth answers. After training with the complementary samples (ie, the original and generated samples), the VQA models are forced to focus on all critical objects and words, which significantly improves both visual-explainable and question-sensitive abilities. In return, the performance of these models is further boosted. Extensive ablations have shown the effectiveness of CSS. Particularly, by building on top of the model LMH, we achieve a record-breaking performance of 58.95% on VQA-CP v2, with 6.5% gains.

cs.CV↗

arXiv2020

Evaluation Framework For Large-scale Federated Learning

Federated learning is proposed as a machine learning setting to enable distributed edge devices, such as mobile phones, to collaboratively learn a shared prediction model while keeping all the training data on device, which can not only take full advantage of data distributed across millions of nodes to train a good model but also protect data privacy. However, learning in scenario above poses new challenges. In fact, data across a massive number of unreliable devices is likely to be non-IID (identically and independently distributed), which may make the performance of models trained by federated learning unstable. In this paper, we introduce a framework designed for large-scale federated learning which consists of approaches to generating dataset and modular evaluation framework. Firstly, we construct a suite of open-source non-IID datasets by providing three respects including covariate shift, prior probability shift, and concept shift, which are grounded in real-world assumptions. In addition, we design several rigorous evaluation metrics including the number of network nodes, the size of datasets, the number of communication rounds and communication resources etc. Finally, we present an open-source benchmark for large-scale federated learning research.

cs.LG↗

arXiv2020

Reinforcement-Learning based Portfolio Management with Augmented Asset Movement Prediction States

Portfolio management (PM) is a fundamental financial planning task that aims to achieve investment goals such as maximal profits or minimal risks. Its decision process involves continuous derivation of valuable information from various data sources and sequential decision optimization, which is a prospective research direction for reinforcement learning (RL). In this paper, we propose SARL, a novel State-Augmented RL framework for PM. Our framework aims to address two unique challenges in financial PM: (1) data heterogeneity -- the collected information for each asset is usually diverse, noisy and imbalanced (e.g., news articles); and (2) environment uncertainty -- the financial market is versatile and non-stationary. To incorporate heterogeneous data and enhance robustness against environment uncertainty, our SARL augments the asset information with their price movement prediction as additional states, where the prediction can be solely based on financial data (e.g., asset prices) or derived from alternative sources such as news. Experiments on two real-world datasets, (i) Bitcoin market and (ii) HighTech stock market with 7-year Reuters news articles, validate the effectiveness of SARL over existing PM approaches, both in terms of accumulated profits and risk-adjusted profits. Moreover, extensive simulations are conducted to demonstrate the importance of our proposed state augmentation, providing new insights and boosting performance significantly over standard RL-based PM method and other baselines.

q-fin.PM↗

arXiv2020

CIAN: Cross-Image Affinity Net for Weakly Supervised Semantic Segmentation

Weakly supervised semantic segmentation with only image-level labels saves large human effort to annotate pixel-level labels. Cutting-edge approaches rely on various innovative constraints and heuristic rules to generate the masks for every single image. Although great progress has been achieved by these methods, they treat each image independently and do not take account of the relationships across different images. In this paper, however, we argue that the cross-image relationship is vital for weakly supervised segmentation. Because it connects related regions across images, where supplementary representations can be propagated to obtain more consistent and integral regions. To leverage this information, we propose an end-to-end cross-image affinity module, which exploits pixel-level cross-image relationships with only image-level labels. By means of this, our approach achieves 64.3% and 65.3% mIoU on Pascal VOC 2012 validation and test set respectively, which is a new state-of-the-art result by only using image-level labels for weakly supervised semantic segmentation, demonstrating the superiority of our approach.

cs.CV↗

arXiv2019

A low-energy compact Shanghai-Wuhan electron beam ion trap for extraction of highly charged ions

A low-energy, compact and superconducting electron beam ion trap (the Shanghai-Wuhan EBIT or SW-EBIT) for extraction of highly charged ions is presented. The magnetic field in the central drift tube of the SW-EBIT is approximately 0.21 T produced by a pair of high-temperature superconducting coils. The electron-beam energy of the SW-EBIT is in the range of 30-4000 eV, and the maximum electron-beam current is up to 9 mA. Acting as a source of highly charged ions, the ion-beam optics for extraction is integrated, including an ion extractor and an einzel lens. A Wien filter is then used to measure the charge-state distribution of the extracted ions. In this work, the tungsten ions below the charge state of 15 have been produced, extracted, and analyzed. The charge-state distributions and spectra in the range of 530-580 nm of tungsten ions have been measured simultaneously with the electron-beam energy of 279 eV and 300 eV, which preliminarily indicates that the 549.9 nm line comes from $W^{14+}$.

physics.atom-ph↗

arXiv2019

Counterfactual Critic Multi-Agent Training for Scene Graph Generation

Scene graphs -- objects as nodes and visual relationships as edges -- describe the whereabouts and interactions of the things and stuff in an image for comprehensive scene understanding. To generate coherent scene graphs, almost all existing methods exploit the fruitful visual context by modeling message passing among objects, fitting the dynamic nature of reasoning with visual context, eg, "person" on "bike" can help to determine the relationship "ride", which in turn contributes to the category confidence of the two objects. However, we argue that the scene dynamics is not properly learned by using the prevailing cross-entropy based supervised learning paradigm, which is not sensitive to graph inconsistency: errors at the hub or non-hub nodes are unfortunately penalized equally. To this end, we propose a Counterfactual critic Multi-Agent Training (CMAT) approach to resolve the mismatch. CMAT is a multi-agent policy gradient method that frames objects as cooperative agents, and then directly maximizes a graph-level metric as the reward. In particular, to assign the reward properly to each agent, CMAT uses a counterfactual baseline that disentangles the agent-specific reward by fixing the dynamics of other agents. Extensive validations on the challenging Visual Genome benchmark show that CMAT achieves a state-of-the-art by significant performance gains under various settings and metrics.

cs.CV↗

arXiv2019

Weak Supervision Enhanced Generative Network for Question Generation

Automatic question generation according to an answer within the given passage is useful for many applications, such as question answering system, dialogue system, etc. Current neural-based methods mostly take two steps which extract several important sentences based on the candidate answer through manual rules or supervised neural networks and then use an encoder-decoder framework to generate questions about these sentences. These approaches neglect the semantic relations between the answer and the context of the whole passage which is sometimes necessary for answering the question. To address this problem, we propose the Weak Supervision Enhanced Generative Network (WeGen) which automatically discovers relevant features of the passage given the answer span in a weakly supervised manner to improve the quality of generated questions. More specifically, we devise a discriminator, Relation Guider, to capture the relations between the whole passage and the associated answer and then the Multi-Interaction mechanism is deployed to transfer the knowledge dynamically for our question generation system. Experiments show the effectiveness of our method in both automatic evaluations and human evaluations.

cs.CL↗