arXiv subjects

Qi Wu

Publications and source records attributed to Qi Wu.

At least 235 records · Page 13Linked to original sources

arXiv2024

Invariant Random Forest: Tree-Based Model Solution for OOD Generalization

Out-Of-Distribution (OOD) generalization is an essential topic in machine learning. However, recent research is only focusing on the corresponding methods for neural networks. This paper introduces a novel and effective solution for OOD generalization of decision tree models, named Invariant Decision Tree (IDT). IDT enforces a penalty term with regard to the unstable/varying behavior of a split across different environments during the growth of the tree. Its ensemble version, the Invariant Random Forest (IRF), is constructed. Our proposed method is motivated by a theoretical result under mild conditions, and validated by numerical tests with both synthetic and real datasets. The superior performance compared to non-OOD tree models implies that considering OOD generalization for tree models is absolutely necessary and should be given more attention.

cs.LG↗

arXiv2024

Dynamic CVaR Portfolio Construction with Attention-Powered Generative Factor Learning

The dynamic portfolio construction problem requires dynamic modeling of the joint distribution of multivariate stock returns. To achieve this, we propose a dynamic generative factor model which uses random variable transformation as an implicit way of distribution modeling and relies on the Attention-GRU network for dynamic learning and forecasting. The proposed model captures the dynamic dependence among multivariate stock returns, especially focusing on the tail-side properties. We also propose a two-step iterative algorithm to train the model and then predict the time-varying model parameters, including the time-invariant tail parameters. At each investment date, we can easily simulate new samples from the learned generative model, and we further perform CVaR portfolio optimization with the simulated samples to form a dynamic portfolio strategy. The numerical experiment on stock data shows that our model leads to wiser investments that promise higher reward-risk ratios and present lower tail risks.

q-fin.PM↗

arXiv2024

Hidden charmonium decays of spin-2 partner of $X(3872)$

The Belle collaboration recently reported a promising candidate for the spin-2 $D^*\bar{D}^*$ partner of the $X(3872)$, called the $X_2$ for short, having a mass of $(4014.3 \pm 4.0 \pm 1.5)~\mathrm{MeV}$ and a width of $(4 \pm 11 \pm 6)~\mathrm{MeV} $. Assuming the $X_2$ as a pure molecule of the $D^*\bar{D}^*$, we calculated in detail the hidden charmonium decays of the $X_2 \to J/ψV$ and $X_2\toη_cP$ via the intermediate meson loops, where $V = ρ^0\,,ω$ and $P= π^0\,,η\,,η'$. The results indicate that the decay widths are strongly dependent on the $X_2$ mass. At present center value of the mass $4014.3~\mathrm{MeV}$, the width for the $X_2\to J/ψρ^0$ is predicted to be a few tens of keV, while it is on the order of $10^{2\text{-}3}~\mathrm{keV}$ for the $X_2\to J/ ψω$; the predicted width for the $X_2\to η_c π^0$ is about a few keV, while the widths for $X_2\toη_cη$ and $η_cη'$ are around a few tens and tenths of keV, respectively. We also investigated the dependence of the ratios between these widths on the $X_2$ mass and on the $η$-$η'$ mixing angle, which may be good quantities for experiments. We hope that the present calculations would be checked experimentally in the future.

hep-ph↗

arXiv2023

WebVLN: Vision-and-Language Navigation on Websites

Vision-and-Language Navigation (VLN) task aims to enable AI agents to accurately understand and follow natural language instructions to navigate through real-world environments, ultimately reaching specific target locations. We recognise a promising opportunity to extend VLN to a comparable navigation task that holds substantial significance in our daily lives, albeit within the virtual realm: navigating websites on the Internet. This paper proposes a new task named Vision-and-Language Navigation on Websites (WebVLN), where we use question-based instructions to train an agent, emulating how users naturally browse websites. Unlike the existing VLN task that only pays attention to vision and instruction (language), the WebVLN agent further considers underlying web-specific content like HTML, which could not be seen on the rendered web pages yet contains rich visual and textual information. Toward this goal, we contribute a dataset, WebVLN-v1, and introduce a novel approach called Website-aware VLN Network (WebVLN-Net), which is built upon the foundation of state-of-the-art VLN techniques. Experimental results show that WebVLN-Net outperforms current VLN and web-related navigation methods. We believe that the introduction of the new WebVLN task and its dataset will establish a new dimension within the VLN domain and contribute to the broader vision-and-language research community. The code is available at: https://github.com/WebVLN/WebVLN.

cs.CV↗

arXiv2023

MMBaT: A Multi-task Framework for mmWave-based Human Body Reconstruction and Translation Prediction

Human body reconstruction with Millimeter Wave (mmWave) radar point clouds has gained significant interest due to its ability to work in adverse environments and its capacity to mitigate privacy concerns associated with traditional camera-based solutions. Despite pioneering efforts in this field, two challenges persist. Firstly, raw point clouds contain massive noise points, usually caused by the ambient objects and multi-path effects of Radio Frequency (RF) signals. Recent approaches typically rely on prior knowledge or elaborate preprocessing methods, limiting their applicability. Secondly, even after noise removal, the sparse and inconsistent body-related points pose an obstacle to accurate human body reconstruction. To address these challenges, we introduce mmBaT, a novel multi-task deep learning framework that concurrently estimates the human body and predicts body translations in subsequent frames to extract body-related point clouds. Our method is evaluated on two public datasets that are collected with different radar devices and noise levels. A comprehensive comparison against other state-of-the-art methods demonstrates our method has a superior reconstruction performance and generalization ability from noisy raw data, even when compared to methods provided with body-related point clouds.

cs.CV↗

arXiv2023

The Causal Impact of Credit Lines on Spending Distributions

Consumer credit services offered by e-commerce platforms provide customers with convenient loan access during shopping and have the potential to stimulate sales. To understand the causal impact of credit lines on spending, previous studies have employed causal estimators, based on direct regression (DR), inverse propensity weighting (IPW), and double machine learning (DML) to estimate the treatment effect. However, these estimators do not consider the notion that an individual's spending can be understood and represented as a distribution, which captures the range and pattern of amounts spent across different orders. By disregarding the outcome as a distribution, valuable insights embedded within the outcome distribution might be overlooked. This paper develops a distribution-valued estimator framework that extends existing real-valued DR-, IPW-, and DML-based estimators to distribution-valued estimators within Rubin's causal framework. We establish their consistency and apply them to a real dataset from a large e-commerce platform. Our findings reveal that credit lines positively influence spending across all quantiles; however, as credit lines increase, consumers allocate more to luxuries (higher quantiles) than necessities (lower quantiles).

stat.ME↗

arXiv2023

Context-I2W: Mapping Images to Context-dependent Words for Accurate Zero-Shot Composed Image Retrieval

Different from Composed Image Retrieval task that requires expensive labels for training task-specific models, Zero-Shot Composed Image Retrieval (ZS-CIR) involves diverse tasks with a broad range of visual content manipulation intent that could be related to domain, scene, object, and attribute. The key challenge for ZS-CIR tasks is to learn a more accurate image representation that has adaptive attention to the reference image for various manipulation descriptions. In this paper, we propose a novel context-dependent mapping network, named Context-I2W, for adaptively converting description-relevant Image information into a pseudo-word token composed of the description for accurate ZS-CIR. Specifically, an Intent View Selector first dynamically learns a rotation rule to map the identical image to a task-specific manipulation view. Then a Visual Target Extractor further captures local information covering the main targets in ZS-CIR tasks under the guidance of multiple learnable queries. The two complementary modules work together to map an image to a context-dependent pseudo-word token without extra supervision. Our model shows strong generalization ability on four ZS-CIR tasks, including domain conversion, object composition, object manipulation, and attribute manipulation. It obtains consistent and significant performance boosts ranging from 1.88% to 3.60% over the best methods and achieves new state-of-the-art results on ZS-CIR. Our code is available at https://github.com/Pter61/context-i2w.

cs.CV↗

arXiv2023

Beyond ExaBricks: GPU Volume Path Tracing of AMR Data

Adaptive Mesh Refinement (AMR) is becoming a prevalent data representation for scientific visualization. Resulting from large fluid mechanics simulations, the data is usually cell centric, imposing a number of challenges for high quality reconstruction at sample positions. While recent work has concentrated on real-time volume and isosurface rendering on GPUs, the rendering methods used still focus on simple lighting models without scattering events and global illumination. As in other areas of rendering, key to real-time performance are acceleration data structures; in this work we analyze the major bottlenecks of data structures that were originally optimized for camera/primary ray traversal when used with the incoherent ray tracing workload of a volumetric path tracer, and propose strategies to overcome the challenges coming with this.

cs.GR↗

arXiv2023

Continual Self-supervised Learning: Towards Universal Multi-modal Medical Data Representation Learning

Self-supervised learning is an efficient pre-training method for medical image analysis. However, current research is mostly confined to specific-modality data pre-training, consuming considerable time and resources without achieving universality across different modalities. A straightforward solution is combining all modality data for joint self-supervised pre-training, which poses practical challenges. Firstly, our experiments reveal conflicts in representation learning as the number of modalities increases. Secondly, multi-modal data collected in advance cannot cover all real-world scenarios. In this paper, we reconsider versatile self-supervised learning from the perspective of continual learning and propose MedCoSS, a continuous self-supervised learning approach for multi-modal medical data. Unlike joint self-supervised learning, MedCoSS assigns different modality data to different training stages, forming a multi-stage pre-training process. To balance modal conflicts and prevent catastrophic forgetting, we propose a rehearsal-based continual learning method. We introduce the k-means sampling strategy to retain data from previous modalities and rehearse it when learning new modalities. Instead of executing the pretext task on buffer data, a feature distillation strategy and an intra-modal mixup strategy are applied to these data for knowledge retention. We conduct continuous self-supervised pre-training on a large-scale multi-modal unlabeled dataset, including clinical reports, X-rays, CT scans, MRI scans, and pathological images. Experimental results demonstrate MedCoSS's exceptional generalization ability across nine downstream datasets and its significant scalability in integrating new modality data. Code and pre-trained weight are available at https://github.com/yeerwen/MedCoSS.

cs.CV↗

arXiv2023

Watermarking Vision-Language Pre-trained Models for Multi-modal Embedding as a Service

Recent advances in vision-language pre-trained models (VLPs) have significantly increased visual understanding and cross-modal analysis capabilities. Companies have emerged to provide multi-modal Embedding as a Service (EaaS) based on VLPs (e.g., CLIP-based VLPs), which cost a large amount of training data and resources for high-performance service. However, existing studies indicate that EaaS is vulnerable to model extraction attacks that induce great loss for the owners of VLPs. Protecting the intellectual property and commercial ownership of VLPs is increasingly crucial yet challenging. A major solution of watermarking model for EaaS implants a backdoor in the model by inserting verifiable trigger embeddings into texts, but it is only applicable for large language models and is unrealistic due to data and model privacy. In this paper, we propose a safe and robust backdoor-based embedding watermarking method for VLPs called VLPMarker. VLPMarker utilizes embedding orthogonal transformation to effectively inject triggers into the VLPs without interfering with the model parameters, which achieves high-quality copyright verification and minimal impact on model performance. To enhance the watermark robustness, we further propose a collaborative copyright verification strategy based on both backdoor trigger and embedding distribution, enhancing resilience against various attacks. We increase the watermark practicality via an out-of-distribution trigger selection approach, removing access to the model training data and thus making it possible for many real-world scenarios. Our extensive experiments on various datasets indicate that the proposed watermarking approach is effective and safe for verifying the copyright of VLPs for multi-modal EaaS and robust against model extraction attacks. Our code is available at https://github.com/Pter61/vlpmarker.

cs.CR↗

arXiv2023

Deconfined quantum critical point lost in pressurized SrCu2(BO3)2

In the field of correlated electron materials, the relation between the resonating spin singlet and antiferromagnetic states has long been an attractive topic for understanding of the interesting macroscopic quantum phenomena, such as the ones emerging from magnetic frustrated materials, antiferromagnets and high-temperature superconductors. SrCu2(BO3)2 is a well-known quantum magnet, and it is theoretically expected to be the candidate of correlated electron material for clarifying the existence of a pressure-induced deconfined quantum critical point (DQCP), featured by a continuous quantum phase transition, between the plaquette-singlet (PS) valence bond solid phase and the antiferromagnetic (AF) phase. However, the real nature of the transition is yet to be identified experimentally due to the technical challenge. Here we show the experimental results for the first time, through the state-of-the-art high-pressure heat capacity measurement, that the PS-AF phase transition of the pressurized SrCu2(BO3)2 at zero field is clearly a first-order one. Our result clarifies the more than two-decade long debates about this key issue, and resonates nicely with the recent quantum entanglement understanding that the theoretically predicted DQCPs in representative lattice models are actually a first-order transition. Intriguingly, we also find that the transition temperatures of the PS and AF phase meet at the same pressure-temperature point, which signifies a bi-critical point as those observed in Fe-based superconductor and heavy-fermion compound, and constitutes the first experimental discovery of the pressure-induced bi-critical point in frustrated magnets. Our results provide fresh information for understanding the evolution among different spin states of correlated electron materials under pressure.

cond-mat.str-el↗

arXiv2023

NavGPT: Explicit Reasoning in Vision-and-Language Navigation with Large Language Models

Trained with an unprecedented scale of data, large language models (LLMs) like ChatGPT and GPT-4 exhibit the emergence of significant reasoning abilities from model scaling. Such a trend underscored the potential of training LLMs with unlimited language data, advancing the development of a universal embodied agent. In this work, we introduce the NavGPT, a purely LLM-based instruction-following navigation agent, to reveal the reasoning capability of GPT models in complex embodied scenes by performing zero-shot sequential action prediction for vision-and-language navigation (VLN). At each step, NavGPT takes the textual descriptions of visual observations, navigation history, and future explorable directions as inputs to reason the agent's current status, and makes the decision to approach the target. Through comprehensive experiments, we demonstrate NavGPT can explicitly perform high-level planning for navigation, including decomposing instruction into sub-goal, integrating commonsense knowledge relevant to navigation task resolution, identifying landmarks from observed scenes, tracking navigation progress, and adapting to exceptions with plan adjustment. Furthermore, we show that LLMs is capable of generating high-quality navigational instructions from observations and actions along a path, as well as drawing accurate top-down metric trajectory given the agent's navigation history. Despite the performance of using NavGPT to zero-shot R2R tasks still falling short of trained models, we suggest adapting multi-modality inputs for LLMs to use as visual navigation agents and applying the explicit reasoning of LLMs to benefit learning-based models.

cs.CV↗

arXiv2023

Pionic transitions from $Z_c(4020)$ to $D$ wave charmonia

In the present work, we investigate the charmed meson loops contributions to the pionic transitions from $Z_c(4020)^+$ to the $D$ wave triplets charmonia by using an effective Lagrangian approach. Our estimations indicate that the predicted branching fraction of $Z_c(4020)^+ \to π^+ ψ(1^3D_J) , \ J=(1,2,3)$ are much smaller than the one of $Z_c(4020)^+ \to π^+ h_c $. Thus, searching $Z_c(4020)^\pm$ in the $π^\pm ψ(1^3D_J) $ invariant mass distributions is impossible. Thus, the observed peak structures at 4.04 and 4.13 GeV in the $π^\pm ψ(3770)$ invariant mass distributions should not come from the contributions of $Z_c(4020)^\pm$, and further precise experimental measurements of the $e^+ e^- \to π^+ π^- ψ(3770)$ process are needed to decode the nature of these two peak structures.

hep-ph↗

arXiv2023

Align before Search: Aligning Ads Image to Text for Accurate Cross-Modal Sponsored Search

Cross-Modal sponsored search displays multi-modal advertisements (ads) when consumers look for desired products by natural language queries in search engines. Since multi-modal ads bring complementary details for query-ads matching, the ability to align ads-specific information in both images and texts is crucial for accurate and flexible sponsored search. Conventional research mainly studies from the view of modeling the implicit correlations between images and texts for query-ads matching, ignoring the alignment of detailed product information and resulting in suboptimal search performance.In this work, we propose a simple alignment network for explicitly mapping fine-grained visual parts in ads images to the corresponding text, which leverages the co-occurrence structure consistency between vision and language spaces without requiring expensive labeled training data. Moreover, we propose a novel model for cross-modal sponsored search that effectively conducts the cross-modal alignment and query-ads matching in two separate processes. In this way, the model matches the multi-modal input in the same language space, resulting in a superior performance with merely half of the training data. Our model outperforms the state-of-the-art models by 2.57% on a large commercial dataset. Besides sponsored search, our alignment method is applicable for general cross-modal search. We study a typical cross-modal retrieval task on the MSCOCO dataset, which achieves consistent performance improvement and proves the generalization ability of our method. Our code is available at https://github.com/Pter61/AlignCMSS/

cs.CV↗

arXiv2023

SwitchGPT: Adapting Large Language Models for Non-Text Outputs

Large Language Models (LLMs), primarily trained on text-based datasets, exhibit exceptional proficiencies in understanding and executing complex linguistic instructions via text outputs. However, they falter when requests to generate non-text ones. Concurrently, modality conversion models, such as text-to-image, despite generating high-quality images, suffer from a lack of extensive textual pretraining. As a result, these models are only capable of accommodating specific image descriptions rather than comprehending more complex instructions. To bridge this gap, we propose a novel approach, \methodname, from a modality conversion perspective that evolves a text-based LLM into a multi-modal one. We specifically employ a minimal dataset to instruct LLMs to recognize the intended output modality as directed by the instructions. Consequently, the adapted LLM can effectively summon various off-the-shelf modality conversion models from the model zoos to generate non-text responses. This circumvents the necessity for complicated pretraining that typically requires immense quantities of paired multi-modal data, while simultaneously inheriting the extensive knowledge of LLMs and the ability of high-quality generative models. To evaluate and compare the adapted multi-modal LLM with its traditional counterparts, we have constructed a multi-modal instruction benchmark that solicits diverse modality outputs. The experiment results reveal that, with minimal training, LLMs can be conveniently adapted to comprehend requests for non-text responses, thus achieving higher flexibility in multi-modal scenarios. Code and data will be made available at https://github.com/xinke-wang/SwitchGPT.

cs.CV↗

arXiv2023

S3C: Semi-Supervised VQA Natural Language Explanation via Self-Critical Learning

VQA Natural Language Explanation (VQA-NLE) task aims to explain the decision-making process of VQA models in natural language. Unlike traditional attention or gradient analysis, free-text rationales can be easier to understand and gain users' trust. Existing methods mostly use post-hoc or self-rationalization models to obtain a plausible explanation. However, these frameworks are bottlenecked by the following challenges: 1) the reasoning process cannot be faithfully responded to and suffer from the problem of logical inconsistency. 2) Human-annotated explanations are expensive and time-consuming to collect. In this paper, we propose a new Semi-Supervised VQA-NLE via Self-Critical Learning (S3C), which evaluates the candidate explanations by answering rewards to improve the logical consistency between answers and rationales. With a semi-supervised learning framework, the S3C can benefit from a tremendous amount of samples without human-annotated explanations. A large number of automatic measures and human evaluations all show the effectiveness of our method. Meanwhile, the framework achieves a new state-of-the-art performance on the two VQA-NLE datasets.

cs.CV↗

arXiv2023

DeLELSTM: Decomposition-based Linear Explainable LSTM to Capture Instantaneous and Long-term Effects in Time Series

Time series forecasting is prevalent in various real-world applications. Despite the promising results of deep learning models in time series forecasting, especially the Recurrent Neural Networks (RNNs), the explanations of time series models, which are critical in high-stakes applications, have received little attention. In this paper, we propose a Decomposition-based Linear Explainable LSTM (DeLELSTM) to improve the interpretability of LSTM. Conventionally, the interpretability of RNNs only concentrates on the variable importance and time importance. We additionally distinguish between the instantaneous influence of new coming data and the long-term effects of historical data. Specifically, DeLELSTM consists of two components, i.e., standard LSTM and tensorized LSTM. The tensorized LSTM assigns each variable with a unique hidden state making up a matrix $\mathbf{h}_t$, and the standard LSTM models all the variables with a shared hidden state $\mathbf{H}_t$. By decomposing the $\mathbf{H}_t$ into the linear combination of past information $\mathbf{h}_{t-1}$ and the fresh information $\mathbf{h}_{t}-\mathbf{h}_{t-1}$, we can get the instantaneous influence and the long-term effect of each variable. In addition, the advantage of linear regression also makes the explanation transparent and clear. We demonstrate the effectiveness and interpretability of DeLELSTM on three empirical datasets. Extensive experiments show that the proposed method achieves competitive performance against the baseline methods and provides a reliable explanation relative to domain knowledge.

cs.LG↗

arXiv2023

BHSD: A 3D Multi-Class Brain Hemorrhage Segmentation Dataset

Intracranial hemorrhage (ICH) is a pathological condition characterized by bleeding inside the skull or brain, which can be attributed to various factors. Identifying, localizing and quantifying ICH has important clinical implications, in a bleed-dependent manner. While deep learning techniques are widely used in medical image segmentation and have been applied to the ICH segmentation task, existing public ICH datasets do not support the multi-class segmentation problem. To address this, we develop the Brain Hemorrhage Segmentation Dataset (BHSD), which provides a 3D multi-class ICH dataset containing 192 volumes with pixel-level annotations and 2200 volumes with slice-level annotations across five categories of ICH. To demonstrate the utility of the dataset, we formulate a series of supervised and semi-supervised ICH segmentation tasks. We provide experimental results with state-of-the-art models as reference benchmarks for further model developments and evaluations on this dataset.

cs.CV↗