arXiv subjects

Qi Wu

Publications and source records attributed to Qi Wu.

At least 307 records · Page 17Linked to original sources

arXiv2022

Diagnosing Vision-and-Language Navigation: What Really Matters

Vision-and-language navigation (VLN) is a multimodal task where an agent follows natural language instructions and navigates in visual environments. Multiple setups have been proposed, and researchers apply new model architectures or training techniques to boost navigation performance. However, there still exist non-negligible gaps between machines' performance and human benchmarks. Moreover, the agents' inner mechanisms for navigation decisions remain unclear. To the best of our knowledge, how the agents perceive the multimodal input is under-studied and needs investigation. In this work, we conduct a series of diagnostic experiments to unveil agents' focus during navigation. Results show that indoor navigation agents refer to both object and direction tokens when making decisions. In contrast, outdoor navigation agents heavily rely on direction tokens and poorly understand the object tokens. Transformer-based agents acquire a better cross-modal understanding of objects and display strong numerical reasoning ability than non-Transformer-based agents. When it comes to vision-and-language alignments, many models claim that they can align object tokens with specific visual targets. We find unbalanced attention on the vision and text input and doubt the reliability of such cross-modal alignments.

cs.CV↗

arXiv2022

BSRT: Improving Burst Super-Resolution with Swin Transformer and Flow-Guided Deformable Alignment

This work addresses the Burst Super-Resolution (BurstSR) task using a new architecture, which requires restoring a high-quality image from a sequence of noisy, misaligned, and low-resolution RAW bursts. To overcome the challenges in BurstSR, we propose a Burst Super-Resolution Transformer (BSRT), which can significantly improve the capability of extracting inter-frame information and reconstruction. To achieve this goal, we propose a Pyramid Flow-Guided Deformable Convolution Network (Pyramid FG-DCN) and incorporate Swin Transformer Blocks and Groups as our main backbone. More specifically, we combine optical flows and deformable convolutions, hence our BSRT can handle misalignment and aggregate the potential texture information in multi-frames more efficiently. In addition, our Transformer-based structure can capture long-range dependency to further improve the performance. The evaluation on both synthetic and real-world tracks demonstrates that our approach achieves a new state-of-the-art in BurstSR task. Further, our BSRT wins the championship in the NTIRE2022 Burst Super-Resolution Challenge.

cs.CV↗

arXiv2022

An Automated FPGA-based Framework for Rapid Prototyping of Nonbinary LDPC Codes

Nonbinary LDPC codes have shown superior performance close to the Shannon limit. Compared to binary LDPC codes of similar lengths, they can reach orders of magnitudes lower error rate. However, multitude of design freedoms of nonbinary LDPC codes complicates the practical code and decoder design process. Fast simulations are critically important to evaluate the pros and cons. Rapid prototyping on FPGA is attractive but takes significant design efforts due to its high design complexity. We propose a high-throughput reconfigurable hardware emulation architecture with decoder and peripheral co-design. The architecture enables a library and script-based framework that automates the construction of FPGA emulations. Code and decoder design parameters are programmed either during run time or by script in design time. We demonstrate the capability of the framework in evaluating practical code and decoder design by experimenting with two popular nonbinary LDPC codes, regular (2, dc) codes and quasi-cyclic codes: each emulation model can be auto-constructed within hours and the decoder delivers excellent error-correcting performance on a Xilinx Virtex-5 FPGA with throughput of up to hundreds of Mbps.

cs.IT↗

arXiv2022

HOP: History-and-Order Aware Pre-training for Vision-and-Language Navigation

Pre-training has been adopted in a few of recent works for Vision-and-Language Navigation (VLN). However, previous pre-training methods for VLN either lack the ability to predict future actions or ignore the trajectory contexts, which are essential for a greedy navigation process. In this work, to promote the learning of spatio-temporal visual-textual correspondence as well as the agent's capability of decision making, we propose a novel history-and-order aware pre-training paradigm (HOP) with VLN-specific objectives that exploit the past observations and support future action prediction. Specifically, in addition to the commonly used Masked Language Modeling (MLM) and Trajectory-Instruction Matching (TIM), we design two proxy tasks to model temporal order information: Trajectory Order Modeling (TOM) and Group Order Modeling (GOM). Moreover, our navigation action prediction is also enhanced by introducing the task of Action Prediction with History (APH), which takes into account the history visual perceptions. Extensive experimental results on four downstream VLN tasks (R2R, REVERIE, NDH, RxR) demonstrate the effectiveness of our proposed method compared against several state-of-the-art agents.

cs.CV↗

arXiv2022

MuKEA: Multimodal Knowledge Extraction and Accumulation for Knowledge-based Visual Question Answering

Knowledge-based visual question answering requires the ability of associating external knowledge for open-ended cross-modal scene understanding. One limitation of existing solutions is that they capture relevant knowledge from text-only knowledge bases, which merely contain facts expressed by first-order predicates or language descriptions while lacking complex but indispensable multimodal knowledge for visual understanding. How to construct vision-relevant and explainable multimodal knowledge for the VQA scenario has been less studied. In this paper, we propose MuKEA to represent multimodal knowledge by an explicit triplet to correlate visual objects and fact answers with implicit relations. To bridge the heterogeneous gap, we propose three objective losses to learn the triplet representations from complementary views: embedding structure, topological relation and semantic space. By adopting a pre-training and fine-tuning learning strategy, both basic and domain-specific multimodal knowledge are progressively accumulated for answer prediction. We outperform the state-of-the-art by 3.35% and 6.08% respectively on two challenging knowledge-required datasets: OK-VQA and KRVQA. Experimental results prove the complementary benefits of the multimodal knowledge with existing knowledge bases and the advantages of our end-to-end framework over the existing pipeline methods. The code is available at https://github.com/AndersonStra/MuKEA.

cs.CV↗

arXiv2022

Bridging the Gap Between Learning in Discrete and Continuous Environments for Vision-and-Language Navigation

Most existing works in vision-and-language navigation (VLN) focus on either discrete or continuous environments, training agents that cannot generalize across the two. The fundamental difference between the two setups is that discrete navigation assumes prior knowledge of the connectivity graph of the environment, so that the agent can effectively transfer the problem of navigation with low-level controls to jumping from node to node with high-level actions by grounding to an image of a navigable direction. To bridge the discrete-to-continuous gap, we propose a predictor to generate a set of candidate waypoints during navigation, so that agents designed with high-level actions can be transferred to and trained in continuous environments. We refine the connectivity graph of Matterport3D to fit the continuous Habitat-Matterport3D, and train the waypoints predictor with the refined graphs to produce accessible waypoints at each time step. Moreover, we demonstrate that the predicted waypoints can be augmented during training to diversify the views and paths, and therefore enhance agent's generalization ability. Through extensive experiments we show that agents navigating in continuous environments with predicted waypoints perform significantly better than agents using low-level actions, which reduces the absolute discrete-to-continuous gap by 11.76% Success Weighted by Path Length (SPL) for the Cross-Modal Matching Agent and 18.24% SPL for the Recurrent VLN-BERT. Our agents, trained with a simple imitation learning objective, outperform previous methods by a large margin, achieving new state-of-the-art results on the testing environments of the R2R-CE and the RxR-CE datasets.

cs.CV↗

arXiv2022

OpenKBP-Opt: An international and reproducible evaluation of 76 knowledge-based planning pipelines

We establish an open framework for developing plan optimization models for knowledge-based planning (KBP) in radiotherapy. Our framework includes reference plans for 100 patients with head-and-neck cancer and high-quality dose predictions from 19 KBP models that were developed by different research groups during the OpenKBP Grand Challenge. The dose predictions were input to four optimization models to form 76 unique KBP pipelines that generated 7600 plans. The predictions and plans were compared to the reference plans via: dose score, which is the average mean absolute voxel-by-voxel difference in dose a model achieved; the deviation in dose-volume histogram (DVH) criterion; and the frequency of clinical planning criteria satisfaction. We also performed a theoretical investigation to justify our dose mimicking models. The range in rank order correlation of the dose score between predictions and their KBP pipelines was 0.50 to 0.62, which indicates that the quality of the predictions is generally positively correlated with the quality of the plans. Additionally, compared to the input predictions, the KBP-generated plans performed significantly better (P<0.05; one-sided Wilcoxon test) on 18 of 23 DVH criteria. Similarly, each optimization model generated plans that satisfied a higher percentage of criteria than the reference plans. Lastly, our theoretical investigation demonstrated that the dose mimicking models generated plans that are also optimal for a conventional planning model. This was the largest international effort to date for evaluating the combination of KBP prediction and optimization models. In the interest of reproducibility, our data and code is freely available at https://github.com/ababier/open-kbp-opt.

physics.med-ph↗

arXiv2021

Data-driven advice for interpreting local and global model predictions in bioinformatics problems

Tree-based algorithms such as random forests and gradient boosted trees continue to be among the most popular and powerful machine learning models used across multiple disciplines. The conventional wisdom of estimating the impact of a feature in tree based models is to measure the \textit{node-wise reduction of a loss function}, which (i) yields only global importance measures and (ii) is known to suffer from severe biases. Conditional feature contributions (CFCs) provide \textit{local}, case-by-case explanations of a prediction by following the decision path and attributing changes in the expected output of the model to each feature along the path. However, Lundberg et al. pointed out a potential bias of CFCs which depends on the distance from the root of a tree. The by now immensely popular alternative, SHapley Additive exPlanation (SHAP) values appear to mitigate this bias but are computationally much more expensive. Here we contribute a thorough comparison of the explanations computed by both methods on a set of 164 publicly available classification problems in order to provide data-driven algorithm recommendations to current researchers. For random forests, we find extremely high similarities and correlations of both local and global SHAP values and CFC scores, leading to very similar rankings and interpretations. Analogous conclusions hold for the fidelity of using global feature importance scores as a proxy for the predictive power associated with each feature.

stat.ML↗

arXiv2021

LocFormer: Enabling Transformers to Perform Temporal Moment Localization on Long Untrimmed Videos With a Feature Sampling Approach

We propose LocFormer, a Transformer-based model for video grounding which operates at a constant memory footprint regardless of the video length, i.e. number of frames. LocFormer is designed for tasks where it is necessary to process the entire long video and at its core lie two main contributions. First, our model incorporates a new sampling technique that splits the input feature sequence into a fixed number of sections and selects a single feature per section using a stochastic approach, which allows us to obtain a feature sample set that is representative of the video content for the task at hand while keeping the memory footprint constant. Second, we propose a modular design that separates functionality, enabling us to learn an inductive bias via supervising the self-attention heads, while also effectively leveraging pre-trained text and video encoders. We test our proposals on relevant benchmark datasets for video grounding, showing that not only LocFormer can achieve excellent results including state-of-the-art performance on YouCookII, but also that our sampling technique is more effective than competing counterparts and that it consistently improves the performance of prior work, by up to 3.13\% in the mean temporal IoU, ultimately leading to a new state-of-the-art performance on Charades-STA.

cs.CV↗

arXiv2021

Adaptive Mimic: Deep Reinforcement Learning of Parameterized Bipedal Walking from Infeasible References

Not until recently, robust robot locomotion has been achieved by deep reinforcement learning (DRL). However, for efficient learning of parametrized bipedal walking, developed references are usually required, limiting the performance to that of the references. In this paper, we propose to design an adaptive reward function for imitation learning from the references. The agent is encouraged to mimic the references when its performance is low, while to pursue high performance when it reaches the limit of references. We further demonstrate that developed references can be replaced by low-quality references that are generated without laborious tuning and infeasible to deploy by themselves, as long as they can provide a priori knowledge to expedite the learning process.

cs.RO↗

arXiv2021

Structured Multimodal Attentions for TextVQA

In this paper, we propose an end-to-end structured multimodal attention (SMA) neural network to mainly solve the first two issues above. SMA first uses a structural graph representation to encode the object-object, object-text and text-text relationships appearing in the image, and then designs a multimodal graph attention network to reason over it. Finally, the outputs from the above modules are processed by a global-local attentional answering module to produce an answer splicing together tokens from both OCR and general vocabulary iteratively by following M4C. Our proposed model outperforms the SoTA models on TextVQA dataset and two tasks of ST-VQA dataset among all models except pre-training based TAP. Demonstrating strong reasoning ability, it also won first place in TextVQA Challenge 2020. We extensively test different OCR methods on several reasoning models and investigate the impact of gradually increased OCR performance on TextVQA benchmark. With better OCR results, different models share dramatic improvement over the VQA accuracy, but our model benefits most blessed by strong textual-visual reasoning ability. To grant our method an upper bound and make a fair testing base available for further works, we also provide human-annotated ground-truth OCR annotations for the TextVQA dataset, which were not given in the original release. The code and ground-truth OCR annotations for the TextVQA dataset are available at https://github.com/ChenyuGAO-CS/SMA

cs.CV↗

arXiv2021

Production of $Z_{cs}$ in $B$ and $B_s$ decay

In the present work, we investigate the production of $Z_{cs}^+$ in $B^+$ and $B_s^0$ decay, where $Z_{cs}^+$ is assigned as a $D_s^{+} \bar{D}^{\ast0} + D_s^{\ast +}\bar{D}^0$ molecular state. By using an effective Lagrangian approach, we evaluate the branching ratio of $B^0_s\rightarrow K^- Z^+_{cs}$ and $B^+\rightarrow ϕZ^{+}_{cs}$ via the triangle loop mechanism. The estimated branching fractions of $B^0_s\rightarrow K^- Z^+_{cs}$ and $B^+\rightarrow ϕZ^{+}_{cs}$ are an order of $10^{-4} $ and $10^{-5}$, respectively. The ratio of these two branching fraction is estimated to be about 5, which indicate that the $B_s^0 \to K^\pm Z^\mp_{cs} \to K^+ K^- J/ψ$ may be a better process of searching $Z_{cs}$ and accessible for further experimental measurement of the Belle II and LHCb collaborations.

hep-ph↗

arXiv2021

V2C: Visual Voice Cloning

Existing Voice Cloning (VC) tasks aim to convert a paragraph text to a speech with desired voice specified by a reference audio. This has significantly boosted the development of artificial speech applications. However, there also exist many scenarios that cannot be well reflected by these VC tasks, such as movie dubbing, which requires the speech to be with emotions consistent with the movie plots. To fill this gap, in this work we propose a new task named Visual Voice Cloning (V2C), which seeks to convert a paragraph of text to a speech with both desired voice specified by a reference audio and desired emotion specified by a reference video. To facilitate research in this field, we construct a dataset, V2C-Animation, and propose a strong baseline based on existing state-of-the-art (SoTA) VC techniques. Our dataset contains 10,217 animated movie clips covering a large variety of genres (e.g., Comedy, Fantasy) and emotions (e.g., happy, sad). We further design a set of evaluation metrics, named MCD-DTW-SL, which help evaluate the similarity between ground-truth speeches and the synthesised ones. Extensive experimental results show that even SoTA VC methods cannot generate satisfying speeches for our V2C task. We hope the proposed new task together with the constructed dataset and evaluation metric will facilitate the research in the field of voice cloning and the broader vision-and-language community.

cs.CV↗

arXiv2021

Risk and return prediction for pricing portfolios of non-performing consumer credit

We design a system for risk-analyzing and pricing portfolios of non-performing consumer credit loans. The rapid development of credit lending business for consumers heightens the need for trading portfolios formed by overdue loans as a manner of risk transferring. However, the problem is nontrivial technically and related research is absent. We tackle the challenge by building a bottom-up architecture, in which we model the distribution of every single loan's repayment rate, followed by modeling the distribution of the portfolio's overall repayment rate. To address the technical issues encountered, we adopt the approaches of simultaneous quantile regression, R-copula, and Gaussian one-factor copula model. To our best knowledge, this is the first study that successfully adopts a bottom-up system for analyzing credit portfolio risks of consumer loans. We conduct experiments on a vast amount of data and prove that our methodology can be applied successfully in real business tasks.

q-fin.RM↗

arXiv2021

Exploration of the hidden charm decays of $Z_{cs}(3985)$

In the present work, we investigate the decays of $Z^-_{cs}(3985)$ in the $D_s^{-} D^{\ast0} + D_s^{\ast -}D^0$ molecular frame. By using an effective Lagrangian approach, the branching ratios of $Z^-_{cs}\rightarrow J/ψK^-$ and $Z^-_{cs}\rightarrow η_c K^{\ast-}$ are estimated. It is found that the dominating decay mode of $Z^-_{cs}(3985)$ is open charm channel, while, the branching ratio of the hidden charm decay channels is about $20\%$. In particular, the branching ratios of $Z^-_{cs}\rightarrow J/ψK^-$ and $Z^-_{cs}\rightarrow η_c K^{\ast-}$ are predicted to be of order of $10\%$. Moreover, our estimation indicates that the fraction of the quasi-two-body cascade decay via $Z_{cs}(3985)$ and the total cross section of $e^+ e^- \to K^+ K^- J/ψ$ around 4.68 GeV is $\left(6.8^{+10.5}_{-6.2} \right)\%$. All the predicted branching ratios in the present work could be tested by further measurements in BESIII and Belle II.

hep-ph↗

arXiv2021

What causes the absence of pulsations in Central Compact Objects in Supernova Remnants?

Most young neutron stars belonging to the class of Central Compact Objects in supernova remnants (CCOs) do not have known periodicities. We investigated seven such CCOs to understand the common reasons for the absence of detected pulsations. Making use of XMM-Newton, Chandra, and NICER observations, we perform a systematic timing and spectral analysis to derive updated sensitivity limits for both periodic signals and multi-temperature spectral components that could be associated with radiation from hotspots on the neutron star surface. Based on these limits, we then investigated for each target the allowed viewing geometry that could explain the lack of pulsations. We estimate it is unlikely ($< 10^{-6}$) to attribute that we do not see pulsations to an unfavorable viewing geometry for five considered sources. Alternatively, the carbon atmosphere model, which assumes homogeneous temperature distribution on the surface, describes the spectra equally well and provides a reasonable interpretation for the absence of detected periodicities within current limits. The unusual properties of CCOs with respect to other young neutron stars could suggest a different evolutionary path, as that proposed for sources experiencing episodes of significant fallback accretion after the supernova event.

astro-ph.HE↗

arXiv2021

The rank of a signed graph in terms of girth

Let $Γ=(G,σ)$ be a signed graph and $A(G,σ)$ be its adjacency matrix. Denote by $gr(G)$ the girth of $G$, which is the length of the shortest cycle in $G$. Let $r(G,σ)$ be the rank of $(G,σ)$. In this paper, we will prove that $r(G,σ)\geq gr(G)-2$ for a signed graph $(G,σ)$. Moreover, we characterize all extremal graphs which satisfy the equalities $r(G,σ)=gr(G)-2$ and $r(G,σ)=gr(G)$.

math.CO↗

arXiv2021

The Road to Know-Where: An Object-and-Room Informed Sequential BERT for Indoor Vision-Language Navigation

Vision-and-Language Navigation (VLN) requires an agent to find a path to a remote location on the basis of natural-language instructions and a set of photo-realistic panoramas. Most existing methods take the words in the instructions and the discrete views of each panorama as the minimal unit of encoding. However, this requires a model to match different nouns (e.g., TV, table) against the same input view feature. In this work, we propose an object-informed sequential BERT to encode visual perceptions and linguistic instructions at the same fine-grained level, namely objects and words. Our sequential BERT also enables the visual-textual clues to be interpreted in light of the temporal context, which is crucial to multi-round VLN tasks. Additionally, we enable the model to identify the relative direction (e.g., left/right/front/back) of each navigable location and the room type (e.g., bedroom, kitchen) of its current and final navigation goal, as such information is widely mentioned in instructions implying the desired next and final locations. We thus enable the model to know-where the objects lie in the images, and to know-where they stand in the scene. Extensive experiments demonstrate the effectiveness compared against several state-of-the-art methods on three indoor VLN tasks: REVERIE, NDH, and R2R. Project repository: https://github.com/YuankaiQi/ORIST

cs.CL↗