arXiv subjects

Xin Liu

Publications and source records attributed to Xin Liu.

At least 595 records · Page 33Linked to original sources

arXiv2023

Motion Matters: Neural Motion Transfer for Better Camera Physiological Measurement

Machine learning models for camera-based physiological measurement can have weak generalization due to a lack of representative training data. Body motion is one of the most significant sources of noise when attempting to recover the subtle cardiac pulse from a video. We explore motion transfer as a form of data augmentation to introduce motion variation while preserving physiological changes of interest. We adapt a neural video synthesis approach to augment videos for the task of remote photoplethysmography (rPPG) and study the effects of motion augmentation with respect to 1) the magnitude and 2) the type of motion. After training on motion-augmented versions of publicly available datasets, we demonstrate a 47% improvement over existing inter-dataset results using various state-of-the-art methods on the PURE dataset. We also present inter-dataset results on five benchmark datasets to show improvements of up to 79% using TS-CAN, a neural rPPG estimation method. Our findings illustrate the usefulness of motion transfer as a data augmentation technique for improving the generalization of models for camera-based physiological sensing. We release our code for using motion transfer as a data augmentation technique on three publicly available datasets, UBFC-rPPG, PURE, and SCAMPS, and models pre-trained on motion-augmented data here: https://motion-matters.github.io/

cs.CV↗

arXiv2023

Static Virus Spread Algorithm for DNA Sequence Design

DNA is not only the genetic material of life, but also a favorable material for a new computing model. Various research works based on DNA computing have been carried out in recent years. DNA sequence design is the foundation of such research. The sequence quality directly affects the universality, robustness, and stability of DNA computing. How to design DNA sequences depends on the biological properties and target requirements, which is a typical combinatorial optimization problem. In this paper, in order to design DNA sequences with high-quality, we propose a novel meta-heuristic evolutionary algorithm, termed the static virus spread algorithm (SVS). Through this algorithm, we focus on the constraints of universal DNA sequence design and produce a large number of DNA sequences with non-complementarity and small difference in melting temperature as the objectives, and fully considering the balanced proportion of the four bases. The computer simulation and polyacrylamide gel electrophoresis experiments show that the high-quality DNA sequences designed by this algorithm are effective, which is expected to provide a convenient tool for sequence preparation before DNA biochemical operations.

cs.ET↗

arXiv2023

IBADR: an Iterative Bias-Aware Dataset Refinement Framework for Debiasing NLU models

As commonly-used methods for debiasing natural language understanding (NLU) models, dataset refinement approaches heavily rely on manual data analysis, and thus maybe unable to cover all the potential biased features. In this paper, we propose IBADR, an Iterative Bias-Aware Dataset Refinement framework, which debiases NLU models without predefining biased features. We maintain an iteratively expanded sample pool. Specifically, at each iteration, we first train a shallow model to quantify the bias degree of samples in the pool. Then, we pair each sample with a bias indicator representing its bias degree, and use these extended samples to train a sample generator. In this way, this generator can effectively learn the correspondence relationship between bias indicators and samples. Furthermore, we employ the generator to produce pseudo samples with fewer biased features by feeding specific bias indicators. Finally, we incorporate the generated pseudo samples into the pool. Experimental results and in-depth analyses on two NLU tasks show that IBADR not only significantly outperforms existing dataset refinement approaches, achieving SOTA, but also is compatible with model-centric methods.

cs.CL↗

arXiv2023

Recaptured Raw Screen Image and Video Demoiréing via Channel and Spatial Modulations

Capturing screen contents by smartphone cameras has become a common way for information sharing. However, these images and videos are often degraded by moiré patterns, which are caused by frequency aliasing between the camera filter array and digital display grids. We observe that the moiré patterns in raw domain is simpler than those in sRGB domain, and the moiré patterns in raw color channels have different properties. Therefore, we propose an image and video demoiréing network tailored for raw inputs. We introduce a color-separated feature branch, and it is fused with the traditional feature-mixed branch via channel and spatial modulations. Specifically, the channel modulation utilizes modulated color-separated features to enhance the color-mixed features. The spatial modulation utilizes the feature with large receptive field to modulate the feature with small receptive field. In addition, we build the first well-aligned raw video demoiréing (RawVDemoiré) dataset and propose an efficient temporal alignment method by inserting alternating patterns. Experiments demonstrate that our method achieves state-of-the-art performance for both image and video demoriéing. We have released the code and dataset in https://github.com/tju-chengyijia/VD_raw.

cs.CV↗

arXiv2023

Open Visual Knowledge Extraction via Relation-Oriented Multimodality Model Prompting

Images contain rich relational knowledge that can help machines understand the world. Existing methods on visual knowledge extraction often rely on the pre-defined format (e.g., sub-verb-obj tuples) or vocabulary (e.g., relation types), restricting the expressiveness of the extracted knowledge. In this work, we take a first exploration to a new paradigm of open visual knowledge extraction. To achieve this, we present OpenVik which consists of an open relational region detector to detect regions potentially containing relational knowledge and a visual knowledge generator that generates format-free knowledge by prompting the large multimodality model with the detected region of interest. We also explore two data enhancement techniques for diversifying the generated format-free visual knowledge. Extensive knowledge quality evaluations highlight the correctness and uniqueness of the extracted open visual knowledge by OpenVik. Moreover, integrating our extracted knowledge across various visual reasoning applications shows consistent improvements, indicating the real-world applicability of OpenVik.

cs.CL↗

arXiv2023

Complex Query Answering on Eventuality Knowledge Graph with Implicit Logical Constraints

Querying knowledge graphs (KGs) using deep learning approaches can naturally leverage the reasoning and generalization ability to learn to infer better answers. Traditional neural complex query answering (CQA) approaches mostly work on entity-centric KGs. However, in the real world, we also need to make logical inferences about events, states, and activities (i.e., eventualities or situations) to push learning systems from System I to System II, as proposed by Yoshua Bengio. Querying logically from an EVentuality-centric KG (EVKG) can naturally provide references to such kind of intuitive and logical inference. Thus, in this paper, we propose a new framework to leverage neural methods to answer complex logical queries based on an EVKG, which can satisfy not only traditional first-order logic constraints but also implicit logical constraints over eventualities concerning their occurrences and orders. For instance, if we know that "Food is bad" happens before "PersonX adds soy sauce", then "PersonX adds soy sauce" is unlikely to be the cause of "Food is bad" due to implicit temporal constraint. To facilitate consistent reasoning on EVKGs, we propose Complex Eventuality Query Answering (CEQA), a more rigorous definition of CQA that considers the implicit logical constraints governing the temporal order and occurrence of eventualities. In this manner, we propose to leverage theorem provers for constructing benchmark datasets to ensure the answers satisfy implicit logical constraints. We also propose a Memory-Enhanced Query Encoding (MEQE) approach to significantly improve the performance of state-of-the-art neural query encoders on the CEQA task.

cs.CL↗

arXiv2023

CAR: Conceptualization-Augmented Reasoner for Zero-Shot Commonsense Question Answering

The task of zero-shot commonsense question answering evaluates models on their capacity to reason about general scenarios beyond those presented in specific datasets. Existing approaches for tackling this task leverage external knowledge from CommonSense Knowledge Bases (CSKBs) by pretraining the model on synthetic QA pairs constructed from CSKBs. In these approaches, negative examples (distractors) are formulated by randomly sampling from CSKBs using fairly primitive keyword constraints. However, two bottlenecks limit these approaches: the inherent incompleteness of CSKBs limits the semantic coverage of synthetic QA pairs, and the lack of human annotations makes the sampled negative examples potentially uninformative and contradictory. To tackle these limitations above, we propose Conceptualization-Augmented Reasoner (CAR), a zero-shot commonsense question-answering framework that fully leverages the power of conceptualization. Specifically, CAR abstracts a commonsense knowledge triple to many higher-level instances, which increases the coverage of CSKB and expands the ground-truth answer space, reducing the likelihood of selecting false-negative distractors. Extensive experiments demonstrate that CAR more robustly generalizes to answering questions about zero-shot commonsense scenarios than existing methods, including large language models, such as GPT3.5 and ChatGPT. Our codes, data, and model checkpoints are available at https://github.com/HKUST-KnowComp/CAR.

cs.CL↗

arXiv2023

Gold: A Global and Local-aware Denoising Framework for Commonsense Knowledge Graph Noise Detection

Commonsense Knowledge Graphs (CSKGs) are crucial for commonsense reasoning, yet constructing them through human annotations can be costly. As a result, various automatic methods have been proposed to construct CSKG with larger semantic coverage. However, these unsupervised approaches introduce spurious noise that can lower the quality of the resulting CSKG, which cannot be tackled easily by existing denoising algorithms due to the unique characteristics of nodes and structures in CSKGs. To address this issue, we propose Gold (Global and Local-aware Denoising), a denoising framework for CSKGs that incorporates entity semantic information, global rules, and local structural information from the CSKG. Experiment results demonstrate that Gold outperforms all baseline methods in noise detection tasks on synthetic noisy CSKG benchmarks. Furthermore, we show that denoising a real-world CSKG is effective and even benefits the downstream zero-shot commonsense question-answering task.

cs.CL↗

arXiv2023

Diffusion Unit: Interpretable Edge Enhancement and Suppression Learning for 3D Point Cloud Segmentation

3D point clouds are discrete samples of continuous surfaces which can be used for various applications. However, the lack of true connectivity information, i.e., edge information, makes point cloud recognition challenging. Recent edge-aware methods incorporate edge modeling into network designs to better describe local structures. Although these methods show that incorporating edge information is beneficial, how edge information helps remains unclear, making it difficult for users to analyze its usefulness. To shed light on this issue, in this study, we propose a new algorithm called Diffusion Unit (DU) that handles edge information in a principled and interpretable manner while providing decent improvement. First, we theoretically show that DU learns to perform task-beneficial edge enhancement and suppression. Second, we experimentally observe and verify the edge enhancement and suppression behavior. Third, we empirically demonstrate that this behavior contributes to performance improvement. Extensive experiments and analyses performed on challenging benchmarks verify the effectiveness of DU. Specifically, our method achieves state-of-the-art performance in object part segmentation using ShapeNet part and scene segmentation using S3DIS. Our source code is available at https://github.com/martianxiu/DiffusionUnit.

cs.CV↗

arXiv2023

QADYNAMICS: Training Dynamics-Driven Synthetic QA Diagnostic for Zero-Shot Commonsense Question Answering

Zero-shot commonsense Question-Answering (QA) requires models to reason about general situations beyond specific benchmarks. State-of-the-art approaches fine-tune language models on QA pairs constructed from CommonSense Knowledge Bases (CSKBs) to equip the models with more commonsense knowledge in a QA context. However, current QA synthesis protocols may introduce noise from the CSKBs and generate ungrammatical questions and false negative options, which impede the model's ability to generalize. To address these issues, we propose QADYNAMICS, a training dynamics-driven framework for QA diagnostics and refinement. Our approach analyzes the training dynamics of each QA pair at both the question level and option level, discarding machine-detectable artifacts by removing uninformative QA pairs and mislabeled or false-negative options. Extensive experiments demonstrate the effectiveness of our approach, which outperforms all baselines while using only 33% of the synthetic data, even including LLMs such as ChatGPT. Moreover, expert evaluations confirm that our framework significantly improves the quality of QA synthesis. Our codes and model checkpoints are available at https://github.com/HKUST-KnowComp/QaDynamics.

cs.CL↗

arXiv2023

MMTSA: Multimodal Temporal Segment Attention Network for Efficient Human Activity Recognition

Multimodal sensors provide complementary information to develop accurate machine-learning methods for human activity recognition (HAR), but introduce significantly higher computational load, which reduces efficiency. This paper proposes an efficient multimodal neural architecture for HAR using an RGB camera and inertial measurement units (IMUs) called Multimodal Temporal Segment Attention Network (MMTSA). MMTSA first transforms IMU sensor data into a temporal and structure-preserving gray-scale image using the Gramian Angular Field (GAF), representing the inherent properties of human activities. MMTSA then applies a multimodal sparse sampling method to reduce data redundancy. Lastly, MMTSA adopts an inter-segment attention module for efficient multimodal fusion. Using three well-established public datasets, we evaluated MMTSA's effectiveness and efficiency in HAR. Results show that our method achieves superior performance improvements 11.13% of cross-subject F1-score on the MMAct dataset than the previous state-of-the-art (SOTA) methods. The ablation study and analysis suggest that MMTSA's effectiveness in fusing multimodal data for accurate HAR. The efficiency evaluation on an edge device showed that MMTSA achieved significantly better accuracy, lower computational load, and lower inference latency than SOTA methods.

cs.CV↗

arXiv2023

Class-Incremental Learning using Diffusion Model for Distillation and Replay

Class-incremental learning aims to learn new classes in an incremental fashion without forgetting the previously learned ones. Several research works have shown how additional data can be used by incremental models to help mitigate catastrophic forgetting. In this work, following the recent breakthrough in text-to-image generative models and their wide distribution, we propose the use of a pretrained Stable Diffusion model as a source of additional data for class-incremental learning. Compared to competitive methods that rely on external, often unlabeled, datasets of real images, our approach can generate synthetic samples belonging to the same classes as the previously encountered images. This allows us to use those additional data samples not only in the distillation loss but also for replay in the classification loss. Experiments on the competitive benchmarks CIFAR100, ImageNet-Subset, and ImageNet demonstrate how this new approach can be used to further improve the performance of state-of-the-art methods for class-incremental learning on large scale datasets.

cs.LG↗

arXiv2023

Varstrometry for Off-nucleus and Dual Sub-kpc AGN (VODKA): Very Long Baseline Array Searches for Dual or Off-nucleus Quasars and Small-scale Jets

Dual and off-nucleus active supermassive black holes are expected to be common in the hierarchical structure formation paradigm, but their identification at parsec scales remains a challenge due to strict angular resolution requirements. We conduct a systematic study using the Very Long Baseline Array (VLBA) to examine 23 radio-bright candidate dual and off-nucleus quasars. The targets are selected by a novel astrometric technique ("varstrometry") from Gaia, aiming to identify dual or off-nucleus quasars at (sub)kilo-parsec scales. Among these quasars, 8 exhibit either multiple radio components or significant (>3$σ$) positional offsets between the VLBA and Gaia positions. The radio emission from the three candidates which exhibit multiple radio components is likely to originate from small-scale jets based on their morphology. Among the remaining five candidates with significant VLBA-Gaia offsets, three are identified as potential dual quasars at parsec scales, one is likely attributed to small-scale jets, and the origin of the last candidate remains unclear. We explore alternative explanations for the observed VLBA-Gaia offsets. We find no evidence for optical jets at kilo-parsec scales, nor any contamination to Gaia astrometric noise from the host galaxy; misaligned coordinate systems are unlikely to account for our offsets. Our study highlights the promise of the varstrometry technique in discovering candidate dual or off-nucleus quasars and emphasizes the need for further confirmation and investigation to validate and understand these intriguing candidates.

astro-ph.GA↗

arXiv2023

Electrostatic environment and Majorana bound states in full-shell topological insulator nanowires

The combination of a superconductor (SC) and a topological insulator (TI) nanowire was proposed as a potential candidate for realizing Majorana zero modes (MZMs). In this study, we adopt the Schrödinger-Poisson formalism to incorporate the electrostatic environment inside the nanowire and systematically explore its topological properties. Our calculations reveal that the proximity to the SC induces a band bending effect, leading to a non-uniform potential across the TI nanowire. As a consequence, there is an upward shift of the Fermi level within the conduction band. This gives rise to the coexistence of surface and bulk states, localized in an accumulation layer adjacent to the TI-SC interface. When magnetic flux is applied, these occupied states have different flux-penetration areas, suppressing the superconducting gap. However, this impact can be mitigated by increasing the radius of the nanowire. Finally, We demonstrate that MZMs can be achieved across a wide range of parameters centered around one applied flux quantum, $ϕ_0 = h/2e$. Within this regime, MZMs can be realized even in the presence of conduction bands, which are not affected by the band bending effect. These findings provide valuable insights into the practical realization of MZMs in TI nanowire-based devices, especially in the presence of a complicated electrostatic environment.

cond-mat.mes-hall↗

arXiv2023

QUEST: An Efficient Query Evaluation Scheme Towards Scan-Intensive Cross-Model Analysis

Modern data-driven applications require that databases support fast cross-model analytical queries. Achieving fast analytical queries in a database system is challenging since they are usually scan-intensive (i.e., they need to intensively scan over a large number of records) which results in huge I/O and memory costs. And it becomes tougher when the analytical queries are cross-model. It is hard to accelerate cross-model analytical queries in existing databases due to the lack of appropriate storage layout and efficient query processing techniques. In this paper, we present QUEST (QUery Evaluation Scheme Towards scan-intensive cross-model analysis) to push scan-intensive queries down to unified columnar storage layout and seamlessly deliver payloads across different data models. QUEST employs a columnar data layout to unify the representation of multi-model data. Then, a novel index structure named Skip-Tree is developed for QUEST to enable the query evaluation more efficient. With the help of two pairwise bitset-based operations coupled with Skip-Tree, the scan of most irrelevant instances can be pruned so as to avoid the giant intermediate result, thus reducing query response latency and saving the computational resources significantly when evaluating scan-intensive cross-model analysis. The proposed methods are implemented on an open-source platform. Through comprehensive theoretical analysis and extensive experiments, we demonstrate that QUEST improves the performance by 3.7x - 178.2x compared to state-of-the-art multi-model databases when evaluating scan-intensive cross-model analytical queries.

cs.DB↗

arXiv2023

Self-Consistent Narrative Prompts on Abductive Natural Language Inference

Abduction has long been seen as crucial for narrative comprehension and reasoning about everyday situations. The abductive natural language inference ($α$NLI) task has been proposed, and this narrative text-based task aims to infer the most plausible hypothesis from the candidates given two observations. However, the inter-sentential coherence and the model consistency have not been well exploited in the previous works on this task. In this work, we propose a prompt tuning model $α$-PACE, which takes self-consistency and inter-sentential coherence into consideration. Besides, we propose a general self-consistent framework that considers various narrative sequences (e.g., linear narrative and reverse chronology) for guiding the pre-trained language model in understanding the narrative context of input. We conduct extensive experiments and thorough ablation studies to illustrate the necessity and effectiveness of $α$-PACE. The performance of our method shows significant improvement against extensive competitive baselines.

cs.CL↗

arXiv2023

A Survey of Graph Pre-processing Methods: From Algorithmic to Hardware Perspectives

Graph-related applications have experienced significant growth in academia and industry, driven by the powerful representation capabilities of graph. However, efficiently executing these applications faces various challenges, such as load imbalance, random memory access, etc. To address these challenges, researchers have proposed various acceleration systems, including software frameworks and hardware accelerators, all of which incorporate graph pre-processing (GPP). GPP serves as a preparatory step before the formal execution of applications, involving techniques such as sampling, reorder, etc. However, GPP execution often remains overlooked, as the primary focus is directed towards enhancing graph applications themselves. This oversight is concerning, especially considering the explosive growth of real-world graph data, where GPP becomes essential and even dominates system running overhead. Furthermore, GPP methods exhibit significant variations across devices and applications due to high customization. Unfortunately, no comprehensive work systematically summarizes GPP. To address this gap and foster a better understanding of GPP, we present a comprehensive survey dedicated to this area. We propose a double-level taxonomy of GPP, considering both algorithmic and hardware perspectives. Through listing relavent works, we illustrate our taxonomy and conduct a thorough analysis and summary of diverse GPP techniques. Lastly, we discuss challenges in GPP and potential future directions.

cs.AR↗

arXiv2023

Varstrometry for Off-nucleus and Dual sub-Kpc AGN (VODKA). SDSS J1608+2716: A Sub-arcsec Quadruply Lensed Quasar at z=2.575

We report Hubble Space Telescope (HST) Wide Field Camera 3 (WFC3) deep IR (F160W) imaging of SDSS J1608+2716. This system, located at a redshift of z=2.575, was recently reported as a triple quasar candidate with subarcsecond separations ($\sim0.25''$) based on selection from Gaia astrometry and follow-up Keck adaptive optics-assisted integral field unit spectroscopy. Our new HST deep IR imaging reveals the presence of a fourth point-like component located $\sim0.9''$ away from the triple system. Additionally, we detect an edge-on disk galaxy located in between the four point sources. The entire system exhibits a characteristic cusp structure in the context of strong gravitational lensing, and the observed image configuration can be successfully reproduced using a lens model based on a singular isothermal ellipsoid mass profile. These findings indicate that this system is a quadruply lensed quasar. Our results highlight the challenges associated with identifying dual/multiple quasars on $\sim$kpc scales at high redshifts, and emphasize the crucial role of deep, high-resolution IR imaging in robustly confirming such systems.

astro-ph.GA↗