arXiv subjects

Rui Chen

Publications and source records attributed to Rui Chen.

At least 271 records · Page 15Linked to original sources

arXiv2024

Sketch3D: Style-Consistent Guidance for Sketch-to-3D Generation

Recently, image-to-3D approaches have achieved significant results with a natural image as input. However, it is not always possible to access these enriched color input samples in practical applications, where only sketches are available. Existing sketch-to-3D researches suffer from limitations in broad applications due to the challenges of lacking color information and multi-view content. To overcome them, this paper proposes a novel generation paradigm Sketch3D to generate realistic 3D assets with shape aligned with the input sketch and color matching the textual description. Concretely, Sketch3D first instantiates the given sketch in the reference image through the shape-preserving generation process. Second, the reference image is leveraged to deduce a coarse 3D Gaussian prior, and multi-view style-consistent guidance images are generated based on the renderings of the 3D Gaussians. Finally, three strategies are designed to optimize 3D Gaussians, i.e., structural optimization via a distribution transfer mechanism, color optimization with a straightforward MSE loss and sketch similarity optimization with a CLIP-based geometric similarity loss. Extensive visual comparisons and quantitative analysis illustrate the advantage of our Sketch3D in generating realistic 3D assets while preserving consistency with the input.

cs.CV↗

arXiv2024

Near-Visible Topological Edge States in a Silicon Nitride Platform

Demonstrations of topological photonics have so far largely been confined to infrared wavelengths where imaging technology and access to low-dimensional quantum materials are both limited. Here, we designed and fabricated silicon nitride ring-resonator arrays to demonstrate photonic topological edge states at ~780 nm. We observed edge states corresponding to the integer quantum Hall Hamiltonian with topological protection against fabrication disorder. This demonstration extends the concept of topological edge states to the near-visible regime and paves the way for nonlinear and non-Hermitian topological photonics with the rich library of near-visible quantum emitters.

physics.optics↗

arXiv2024

An Optimal Control Framework for Influencing Human Driving Behavior in Mixed-Autonomy Traffic

As autonomous vehicles (AVs) become increasingly prevalent, their interaction with human drivers presents a critical challenge. Current AVs lack social awareness, causing behavior that is often awkward or unsafe. To combat this, social AVs, which are proactive rather than reactive in their behavior, have been explored in recent years. With knowledge of robot-human interaction dynamics, a social AV can influence a human driver to exhibit desired behaviors by strategically altering its own behaviors. In this paper, we present a novel framework for achieving human influence. The foundation of our framework lies in an innovative use of control barrier functions to formulate the desired objectives of influence as constraints in an optimal control problem. The computed controls gradually push the system state toward satisfaction of the objectives, e.g. slowing the human down to some desired speed. We demonstrate the proposed framework's feasibility in a variety of scenarios related to car-following and lane changes, including multi-robot and multi-human configurations. In two case studies, we validate the framework's effectiveness when applied to the problems of traffic flow optimization and aggressive behavior mitigation. Given these results, the main contribution of our framework is its versatility in a wide spectrum of influence objectives and mixed-autonomy configurations.

eess.SY↗

arXiv2024

Real-time Safety Index Adaptation for Parameter-varying Systems via Determinant Gradient Ascend

Safety Index Synthesis (SIS) is critical for deriving safe control laws. Recent works propose to synthesize a safety index (SI) via nonlinear programming and derive a safe control law such that the system 1) achieves forward invariant (FI) with some safe set and 2) guarantees finite time convergence (FTC) to that safe set. However, real-world system dynamics can vary during run-time, making the control law infeasible and invalidating the initial SI. Since the full SIS nonlinear programming is computationally expensive, it is infeasible to re-synthesize the SI each time the dynamics are perturbed. To address that, this paper proposes an efficient approach to adapting the SI to varying system dynamics and maintaining the feasibility of the safe control law. The proposed method leverages determinant gradient ascend and derives a closed-form update to safety index parameters, enabling real-time adaptation performance. A numerical study validates the effectiveness of our approach.

eess.SY↗

arXiv2024

Four-dimensional Floquet topological insulator with an emergent second Chern number

Floquet topological insulators have been widely investigated in lower-dimensional systems. However, Floquet topological insulators induced by time-periodic driving in higher-dimensional systems remain unexplored. In this work, we study the effects of time-periodic driving in a four-dimensional (4D) normal insulator, focusing on topological phase transitions at the resonant quasienergy gap. We consider two types of time-periodic driving, including a time-periodic onsite potential and a time-periodic vector potential. We reveal that both types of time-periodic driving can transform the 4D normal insulator into a 4D Floquet topological insulator characterized by an emergent second Chern number. Moreover, it is found that the topological phase of the 4D system can be modulated by tuning the strength and frequency of the time-periodic driving. Our work will be helpful for the future investigation of Floquet topological insulators in higher dimensions.

cond-mat.mes-hall↗

arXiv2024

SSDRec: Self-Augmented Sequence Denoising for Sequential Recommendation

Traditional sequential recommendation methods assume that users' sequence data is clean enough to learn accurate sequence representations to reflect user preferences. In practice, users' sequences inevitably contain noise (e.g., accidental interactions), leading to incorrect reflections of user preferences. Consequently, some pioneer studies have explored modeling sequentiality and correlations in sequences to implicitly or explicitly reduce noise's influence. However, relying on only available intra-sequence information (i.e., sequentiality and correlations in a sequence) is insufficient and may result in over-denoising and under-denoising problems (OUPs), especially for short sequences. To improve reliability, we propose to augment sequences by inserting items before denoising. However, due to the data sparsity issue and computational costs, it is challenging to select proper items from the entire item universe to insert into proper positions in a target sequence. Motivated by the above observation, we propose a novel framework--Self-augmented Sequence Denoising for sequential Recommendation (SSDRec) with a three-stage learning paradigm to solve the above challenges. In the first stage, we empower SSDRec by a global relation encoder to learn multi-faceted inter-sequence relations in a data-driven manner. These relations serve as prior knowledge to guide subsequent stages. In the second stage, we devise a self-augmentation module to augment sequences to alleviate OUPs. Finally, we employ a hierarchical denoising module in the third stage to reduce the risk of false augmentations and pinpoint all noise in raw sequences. Extensive experiments on five real-world datasets demonstrate the superiority of \model over state-of-the-art denoising methods and its flexible applications to mainstream sequential recommendation models. The source code is available at https://github.com/zc-97/SSDRec.

cs.IR↗

arXiv2024

Low-loss multilevel operation using lossy PCM-integrated silicon photonics

Chalcogenide phase-change materials (PCMs) offer new paradigms for programmable photonic integrated circuits (PICs) thanks to their zero static energy and significant refractive index contrast. However, prototypical PCMs, such as GeSbTe (GST), are lossy in their crystalline phase, albeit transparent in the amorphous state. Moreover, electrically switching PCMs to intermediate states is a stochastic process, limiting programming accuracy. As a result, achieving both low-loss and deterministic multi-level operation with GST remains challenging. Although low-loss PCMs, such as Sb2S3 and Sb2Se3, have been discovered in recent years, they are much less technologically mature. In this work, we propose a design with multiple GST segments to overcome the challenge of deterministic multilevel operation. GST segments are individually controlled by interleaved silicon PIN diode heaters in a binary but reliable fashion, and multiple levels are encoded in their phase sequence. A 1 x 1 programmable unit with two unequal GST segments is experimentally demonstrated, showcasing four distinct operation levels and negligible thermal crosstalk with only one pair of metal contacts. We then extend the design to 1 x 2 and 2 x 2 programmable units. For the 2 x 2 programmable unit design, we propose a phase-detuned three-waveguide directional coupler structure to mitigate the absorption and radiation loss, showing < -1.2 dB loss and three splitting ratios. Our work provides a new path toward low-loss and multi-level optical switches using lossy PCMs.

physics.optics↗

arXiv2024

Adaptive Hardness Negative Sampling for Collaborative Filtering

Negative sampling is essential for implicit collaborative filtering to provide proper negative training signals so as to achieve desirable performance. We experimentally unveil a common limitation of all existing negative sampling methods that they can only select negative samples of a fixed hardness level, leading to the false positive problem (FPP) and false negative problem (FNP). We then propose a new paradigm called adaptive hardness negative sampling (AHNS) and discuss its three key criteria. By adaptively selecting negative samples with appropriate hardnesses during the training process, AHNS can well mitigate the impacts of FPP and FNP. Next, we present a concrete instantiation of AHNS called AHNS_{p<0}, and theoretically demonstrate that AHNS_{p<0} can fit the three criteria of AHNS well and achieve a larger lower bound of normalized discounted cumulative gain. Besides, we note that existing negative sampling methods can be regarded as more relaxed cases of AHNS. Finally, we conduct comprehensive experiments, and the results show that AHNS_{p<0} can consistently and substantially outperform several state-of-the-art competitors on multiple datasets.

cs.IR↗

arXiv2023

Periodically driven four-dimensional topological insulator with tunable second Chern number

In recent years, Floquet engineering has attracted considerable attention as a promising approach for tuning topological phase transitions. In this work, we investigate the effects of high-frequency time-periodic driving in a four-dimensional (4D) topological insulator, focusing on topological phase transitions at the off-resonant quasienergy gap. The 4D topological insulator hosts gapless three-dimensional boundary states characterized by the second Chern number $C_{2}$. We demonstrate that the second Chern number of 4D topological insulators can be modulated by tuning the amplitude of time-periodic driving. This includes transitions from a topological phase with $C_{2}=\pm3$ to another topological phase with $C_{2}=\pm1$, or to a topological phase with an even second Chern number $C_{2}=\pm2$ which is absent in the 4D static system. Finally, the approximation theory in the high-frequency limit further confirms the numerical conclusions.

cond-mat.mes-hall↗

arXiv2023

Enhancing Generalizable 6D Pose Tracking of an In-Hand Object with Tactile Sensing

When manipulating an object to accomplish complex tasks, humans rely on both vision and touch to keep track of the object's 6D pose. However, most existing object pose tracking systems in robotics rely exclusively on visual signals, which hinder a robot's ability to manipulate objects effectively. To address this limitation, we introduce TEG-Track, a tactile-enhanced 6D pose tracking system that can track previously unseen objects held in hand. From consecutive tactile signals, TEG-Track optimizes object velocities from marker flows when slippage does not occur, or regresses velocities using a slippage estimation network when slippage is detected. The estimated object velocities are integrated into a geometric-kinematic optimization scheme to enhance existing visual pose trackers. To evaluate our method and to facilitate future research, we construct a real-world dataset for visual-tactile in-hand object pose tracking. Experimental results demonstrate that TEG-Track consistently enhances state-of-the-art generalizable 6D pose trackers in synthetic and real-world scenarios. Our code and dataset are available at https://github.com/leolyliu/TEG-Track.

cs.CV↗

arXiv2023

Phase transitions in intrinsic magnetic topological insulator with high-frequency pumping

In this work, we investigate the topological phase transitions in an effective model for a topological thin film with high-frequency pumping. In particular, our results show that the circularly polarized light can break the time-reversal symmetry and induce the quantum anomalous Hall insulator (QAHI) phase. Meanwhile, the bulk magnetic moment can also break the time-reversal symmetry. Therefore, it shows rich phase diagram by tunning the intensity of the light and the thickness of the thin film. Using the parameters fitted by experimental data, we give the topological phase diagram of the Cr-doped Bi$_{2}$Se$_{3}$ thin film, showing that by modulating the strength of the polarized optical field in an experimentally accessible range, there are four different phases: the normal insulator phase, the time-reversal-symmetry-broken quantum spin Hall insulator phase, and two different QAHI phases with opposite Chern numbers. Comparing with the non-doped Bi$_{2}$Se$_{3}$, it is found that the interplay between the light and bulk magnetic moment separates the two different QAHI phases with opposite Chern numbers. The results show that an intrinsic magnetic topological insulator with high-frequency pumping is an ideal platform for further exploring various topological phenomena with a spontaneously broken time-reversal symmetry.

cond-mat.mes-hall↗

arXiv2023

Light-induced phase crossovers in a quantum spin Hall system

In this work, we theoretically investigate the light-induced topological phases and finite-size crossovers in a paradigmatic quantum spin Hall (QSH) system with high-frequency pumping optics. Taking the HgTe quantum well for an example, our numerical results show that circularly polarized light can break time-reversal symmetry and induce the quantum anomalous Hall (QAH) phase. In particular, the coupling between the edge states is spin dependent and is related not only to the size of the system, but also to the strength of the polarized pumping optics. By tuning the two parameters (system width and optical pumping strength), we obtain four transport regimes, namely, QSH, QAH, edge conducting, and normal insulator. These four different transport regimes have contrasting edge conducting properties, which will feature prominently in transport experiments on various topological materials.

cond-mat.mes-hall↗

arXiv2023

MMSE Design of RIS-aided Communications

Consider a communication system in which a single antenna user equipment exchanges information with a multi-antenna base station via a reconfigurable intelligent surface (RIS) in the presence of spatially correlated channels and electromagnetic interference (EMI). To exploit the attractive advantages of RIS technology, accurate configuration of its reflecting elements is crucial. In this paper, we use statistical knowledge of channels and EMI to optimize the RIS elements for i) accurate channel estimation and ii) reliable data transmission. In both cases, our goal is to determine the RIS coefficients that minimize the mean square error, resulting in the formulation of two non-convex problems that share the same structure. To solve these two problems, we present an alternating optimization approach that reliably converges to a locally optimal solution. The incorporation of the diagonally scaled steepest descent algorithm, derived from Newton's method, ensures fast convergence with manageable complexity. Numerical results demonstrate the effectiveness of the proposed method under various propagation conditions. Notably, it shows significant advantages over existing alternatives that depend on a sub-optimal configuration of the RIS and are derived on the basis of different criteria.

cs.IT↗

arXiv2023

Nonlocal, Flat Band Meta-optics for Monolithic, High Efficiency, Compact Photodetectors

Miniaturized photodetectors are becoming increasingly sought-after components for a range of next generation technologies, such as autonomous vehicles, integrated wearable devices, or gadgets embedded in the Internet of Things. A major challenge, however, lies in shrinking the device footprint, while maintaining high efficiency. This conundrum can be solved by realizing non-trivial relation between the energy and momentum of photons, such as dispersion-free angle-independent devices, known as flat bands. Here, we leverage flat band meta-optics to simultaneously achieve critical absorption over a wide range of incidence angles. For a monolithic silicon meta-optical photodiode, we achieved ~10-fold enhancement in the photon-to-electron conversion efficiency. Such enhancement over a large angular range of ~36 degrees allows incoming light to be collected via a large aperture lens and focused on a compact photodiode, potentially enabling high-speed and low-light operation. Our research unveils new possibilities for creating compact and efficient optoelectronic devices with far-reaching impact on various applications, including augmented reality and light detection and ranging.

physics.optics↗

arXiv2023

Minimum Snap Trajectory Generation and Control for an Under-actuated Flapping Wing Aerial Vehicle

Minimum Snap Trajectory Generation and Control for an Under-actuated Flapping Wing Aerial VehicleThis paper presents both the trajectory generation and tracking control strategies for an underactuated flapping wing aerial vehicle (FWAV). First, the FWAV dynamics is analyzed in a practical perspective. Then, based on these analyses, we demonstrate the differential flatness of the FWAV system, and develop a general-purpose trajectory generation strategy. Subsequently, the trajectory tracking controller is developed with the help of robust control and switch control techniques. After that, the overall system asymptotic stability is guaranteed by Lyapunov stability analysis. To make the controller applicable in real flight, we also provide several instructions. Finally, a series of experiment results manifest the successful implementation of the proposed trajectory generation strategy and tracking control strategy. This work firstly achieves the closed-loop integration of trajectory generation and control for real 3-dimensional flight of an underactuated FWAV to a practical level.

eess.SY↗

arXiv2023

Setting the Trap: Capturing and Defeating Backdoors in Pretrained Language Models through Honeypots

In the field of natural language processing, the prevalent approach involves fine-tuning pretrained language models (PLMs) using local samples. Recent research has exposed the susceptibility of PLMs to backdoor attacks, wherein the adversaries can embed malicious prediction behaviors by manipulating a few training samples. In this study, our objective is to develop a backdoor-resistant tuning procedure that yields a backdoor-free model, no matter whether the fine-tuning dataset contains poisoned samples. To this end, we propose and integrate a honeypot module into the original PLM, specifically designed to absorb backdoor information exclusively. Our design is motivated by the observation that lower-layer representations in PLMs carry sufficient backdoor features while carrying minimal information about the original tasks. Consequently, we can impose penalties on the information acquired by the honeypot module to inhibit backdoor creation during the fine-tuning process of the stem network. Comprehensive experiments conducted on benchmark datasets substantiate the effectiveness and robustness of our defensive strategy. Notably, these results indicate a substantial reduction in the attack success rate ranging from 10\% to 40\% when compared to prior state-of-the-art methods.

cs.LG↗

arXiv2023

Charmoniumlike resonant explanation on the newly observed $X(3960)$

Stimulated by the observation of the newly $X(3960)$ observed by the LHCb collaboration, we adopt the one-boson-exchange model and consider the $S-D$ wave mixing effects to study the $D_s\bar{D}_s/D^*\bar{D}^*/D_s^*\bar{D}_s^*$ interactions with $I(J^{PC})=0(0^{++})$. After producing the phase shifts of this coupled channel systems, our results show that there can exist a charmoniumlike resonance, whose obtained mass and width can both well match with the experimental data of the newly observed $X(3960)$. We also find that the $D^*\bar{D}^*$ system plays an important role in the formation of the newly observed $X(3960)$ as a charmoniumlike resonance, and the $D_s^*\bar{D}_s^*$ system makes a significant contribution to the resonant width. As a byproduct, we perform a coupled channel analysis on the $D^*\bar{D}^*/D_s\bar{D}_s^*/D_s^*\bar{D}_s^*$ interactions with $I(J^{PC})=0(1^{+-})$, our results can predict the existence of the $D_s\bar{D}_s^*$ molecule with $1^{+-}$ and the $D_s^*\bar{D}_s^*$ molecule with $1^{+-}$. Their widths are around several and several to several tens MeV, respectively. Experimental searches for these two possible charmoniumlike molecular candidates can be helpful to verify our proposal.

hep-ph↗

arXiv2023

Chasing Fairness Under Distribution Shift: A Model Weight Perturbation Approach

Fairness in machine learning has attracted increasing attention in recent years. The fairness methods improving algorithmic fairness for in-distribution data may not perform well under distribution shifts. In this paper, we first theoretically demonstrate the inherent connection between distribution shift, data perturbation, and model weight perturbation. Subsequently, we analyze the sufficient conditions to guarantee fairness (i.e., low demographic parity) for the target dataset, including fairness for the source dataset, and low prediction difference between the source and target datasets for each sensitive attribute group. Motivated by these sufficient conditions, we propose robust fairness regularization (RFR) by considering the worst case within the model weight perturbation ball for each sensitive attribute group. We evaluate the effectiveness of our proposed RFR algorithm on synthetic and real distribution shifts across various datasets. Experimental results demonstrate that RFR achieves better fairness-accuracy trade-off performance compared with several baselines. The source code is available at \url{https://github.com/zhimengj0326/RFR_NeurIPS23}.

cs.LG↗