Search

We present a robust refinement method for estimating oriented normals from unstructured point clouds. In contrast to previous approaches that either suffer from high computational complexity or fail to achieve desirable accuracy, our novel framework incorporates sign orientation and data augmentation in the feature space to refine the initial oriented normals, striking a balance between efficiency and accuracy. To address the issue of noise-caused direction inconsistency existing in previous approaches, we introduce a new metric called the Chamfer Normal Distance, which faithfully minimizes the estimation error by correcting the annotated normal with the closest point found on the potentially clean point cloud. This metric not only tackles the challenge but also aids in network training and significantly enhances network robustness against noise. Moreover, we propose an innovative dual-parallel architecture that integrates Multi-scale Local Feature Aggregation and Hierarchical Geometric Information Fusion, which enables the network to capture intricate geometric details more effectively and notably reduces ambiguity in scale selection. Extensive experiments demonstrate the superiority and versatility of our method in both unoriented and oriented normal estimation tasks across synthetic and real-world datasets among indoor and outdoor scenarios. The code is available at https://github.com/YingruiWoo/OCMG-Net.git., Comment: 18 pages, 16 figures

This paper presents a novel non-rigid point set registration method that is inspired by unsupervised clustering analysis. Unlike previous approaches that treat the source and target point sets as separate entities, we develop a holistic framework where they are formulated as clustering centroids and clustering members, separately. We then adopt Tikhonov regularization with an $\ell_1$-induced Laplacian kernel instead of the commonly used Gaussian kernel to ensure smooth and more robust displacement fields. Our formulation delivers closed-form solutions, theoretical guarantees, independence from dimensions, and the ability to handle large deformations. Subsequently, we introduce a clustering-improved Nystr\"om method to effectively reduce the computational complexity and storage of the Gram matrix to linear, while providing a rigorous bound for the low-rank approximation. Our method achieves high accuracy results across various scenarios and surpasses competitors by a significant margin, particularly on shapes with substantial deformations. Additionally, we demonstrate the versatility of our method in challenging tasks such as shape transfer and medical registration., Comment: [CVPR 2024 Highlight] Project and code at: https://github.com/zikai1/CVPR24_PointSetReg

Aligning large language models (LLMs) with human preference has recently gained tremendous attention, with the canonical yet costly RLHF-PPO and the simple and straightforward Direct Preference Optimization (DPO) as two examples. Despite the efficiency, DPO has rarely be used in the state-of-the-art production-level LLMs, implying its potential pathologies. In this work, we revisit DPO with a comprehensive examination of its empirical efficacy and a systematic comparison with RLHF-PPO. We identify the \textbf{3D}-properties of DPO's learning outcomes: the \textbf{D}rastic drop in the likelihood of rejected responses, the \textbf{D}egradation into LLM unlearning, and the \textbf{D}ispersion effect on unseen responses through experiments with both a carefully designed toy model and practical LLMs on tasks including mathematical problem-solving and instruction following. These findings inherently connect to some observations made by related works and we additionally contribute a plausible theoretical explanation for them. Accordingly, we propose easy regularization methods to mitigate the issues caused by \textbf{3D}-properties, improving the training stability and final performance of DPO. Our contributions also include an investigation into how the distribution of the paired preference data impacts the effectiveness of DPO. We hope this work could offer research directions to narrow the gap between reward-free preference learning methods and reward-based ones.

Point cloud normal estimation is a fundamental task in 3D geometry processing. While recent learning-based methods achieve notable advancements in normal prediction, they often overlook the critical aspect of equivariance. This results in inefficient learning of symmetric patterns. To address this issue, we propose E3-Net to achieve equivariance for normal estimation. We introduce an efficient random frame method, which significantly reduces the training resources required for this task to just 1/8 of previous work and improves the accuracy. Further, we design a Gaussian-weighted loss function and a receptive-aware inference strategy that effectively utilizes the local properties of point clouds. Our method achieves superior results on both synthetic and real-world datasets, and outperforms current state-of-the-art techniques by a substantial margin. We improve RMSE by 4% on the PCPNet dataset, 2.67% on the SceneNN dataset, and 2.44% on the FamousShape dataset.

This paper presents an in-depth examination of the evolution and interplay of cognitive and expressive capabilities in large language models (LLMs), with a specific focus on Baichuan-7B and Baichuan-33B, an advanced bilingual (Chinese and English) LLM series. We define and explore the model's cognitive and expressive capabilities through linear representations across three critical phases: Pretraining, Supervised Fine-Tuning (SFT), and Reinforcement Learning from Human Feedback (RLHF). Cognitive capability is defined as the quantity and quality of information conveyed by the neuron output vectors within the network, similar to the neural signal processing in human cognition. Expressive capability is defined as the model's capability to produce word-level output. Our findings unveil a sequential development pattern, where cognitive abilities are largely established during Pretraining, whereas expressive abilities predominantly advance during SFT and RLHF. Statistical analyses confirm a significant correlation between the two capabilities, suggesting that cognitive capacity may limit expressive potential. The paper also explores the theoretical underpinnings of these divergent developmental trajectories and their connection to the LLMs' architectural design. Moreover, we evaluate various optimization-independent strategies, such as few-shot learning and repeated sampling, which bridge the gap between cognitive and expressive capabilities. This research reveals the potential connection between the hidden space and the output space, contributing valuable insights into the interpretability and controllability of their training processes., Comment: Published in ICML 2024

Human preference alignment is critical in building powerful and reliable large language models (LLMs). However, current methods either ignore the multi-dimensionality of human preferences (e.g. helpfulness and harmlessness) or struggle with the complexity of managing multiple reward models. To address these issues, we propose Sequential Preference Optimization (SPO), a method that sequentially fine-tunes LLMs to align with multiple dimensions of human preferences. SPO avoids explicit reward modeling, directly optimizing the models to align with nuanced human preferences. We theoretically derive closed-form optimal SPO policy and loss function. Gradient analysis is conducted to show how SPO manages to fine-tune the LLMs while maintaining alignment on previously optimized dimensions. Empirical results on LLMs of different size and multiple evaluation datasets demonstrate that SPO successfully aligns LLMs across multiple dimensions of human preferences and significantly outperforms the baselines.

Multimodal Sentiment Analysis (MSA) endeavors to understand human sentiment by leveraging language, visual, and acoustic modalities. Despite the remarkable performance exhibited by previous MSA approaches, the presence of inherent multimodal heterogeneities poses a challenge, with the contribution of different modalities varying considerably. Past research predominantly focused on improving representation learning techniques and feature fusion strategies. However, many of these efforts overlooked the variation in semantic richness among different modalities, treating each modality uniformly. This approach may lead to underestimating the significance of strong modalities while overemphasizing the importance of weak ones. Motivated by these insights, we introduce a Text-oriented Cross-Attention Network (TCAN), emphasizing the predominant role of the text modality in MSA. Specifically, for each multimodal sample, by taking unaligned sequences of the three modalities as inputs, we initially allocate the extracted unimodal features into a visual-text and an acoustic-text pair. Subsequently, we implement self-attention on the text modality and apply text-queried cross-attention to the visual and acoustic modalities. To mitigate the influence of noise signals and redundant features, we incorporate a gated control mechanism into the framework. Additionally, we introduce unimodal joint learning to gain a deeper understanding of homogeneous emotional tendencies across diverse modalities through backpropagation. Experimental results demonstrate that TCAN consistently outperforms state-of-the-art MSA methods on two datasets (CMU-MOSI and CMU-MOSEI).

Existing alignment methods share a common topology of information flow, where reward information is collected from humans, modeled with preference learning, and used to tune language models. However, this shared topology has not been systematically characterized, nor have its alternatives been thoroughly explored, leaving the problems of low data efficiency and unreliable generalization unaddressed. As a solution, we introduce a theoretical framework for investigating reward generalization in reinforcement learning from human feedback (RLHF), focusing on the topology of information flow at both macro and micro levels. At the macro level, we portray the RLHF information flow as an autoencoding process over behavior distributions, formalizing the RLHF objective of distributional consistency between human preference and model behavior. At the micro level, we present induced Bayesian networks as a theory of reward generalization in RLHF, introducing fine-grained dataset topologies into generalization bounds. Combining analysis on both levels, we propose reward modeling from tree-structured preference information. It is shown to reduce reward uncertainty by up to $\Theta(\log n/\log\log n)$ times compared to baselines, where $n$ is the dataset size. Validation on three NLP tasks shows that our tree-based reward model achieves an average win rate of 65% against baseline methods, thus improving reward generalization for free via topology design.

Image and video inpainting is a classic problem in computer vision and computer graphics, aiming to fill in the plausible and realistic content in the missing areas of images and videos. With the advance of deep learning, this problem has achieved significant progress recently. The goal of this paper is to comprehensively review the deep learning-based methods for image and video inpainting. Specifically, we sort existing methods into different categories from the perspective of their high-level inpainting pipeline, present different deep learning architectures, including CNN, VAE, GAN, diffusion models, etc., and summarize techniques for module design. We review the training objectives and the common benchmark datasets. We present evaluation metrics for low-level pixel and high-level perceptional similarity, conduct a performance evaluation, and discuss the strengths and weaknesses of representative inpainting methods. We also discuss related real-world applications. Finally, we discuss open challenges and suggest potential future research directions., Comment: accepted to IJCV

This work presents an accurate and robust method for estimating normals from point clouds. In contrast to predecessor approaches that minimize the deviations between the annotated and the predicted normals directly, leading to direction inconsistency, we first propose a new metric termed Chamfer Normal Distance to address this issue. This not only mitigates the challenge but also facilitates network training and substantially enhances the network robustness against noise. Subsequently, we devise an innovative architecture that encompasses Multi-scale Local Feature Aggregation and Hierarchical Geometric Information Fusion. This design empowers the network to capture intricate geometric details more effectively and alleviate the ambiguity in scale selection. Extensive experiments demonstrate that our method achieves the state-of-the-art performance on both synthetic and real-world datasets, particularly in scenarios contaminated by noise. Our implementation is available at https://github.com/YingruiWoo/CMG-Net_Pytorch., Comment: Accepted by AAAI 2024

Large language models (LLMs) have demonstrated remarkable performance on a variety of natural language tasks based on just a few examples of natural language instructions, reducing the need for extensive feature engineering. However, most powerful LLMs are closed-source or limited in their capability for languages other than English. In this technical report, we present Baichuan 2, a series of large-scale multilingual language models containing 7 billion and 13 billion parameters, trained from scratch, on 2.6 trillion tokens. Baichuan 2 matches or outperforms other open-source models of similar size on public benchmarks like MMLU, CMMLU, GSM8K, and HumanEval. Furthermore, Baichuan 2 excels in vertical domains such as medicine and law. We will release all pre-training model checkpoints to benefit the research community in better understanding the training dynamics of Baichuan 2., Comment: Baichuan 2 technical report. Github: https://github.com/baichuan-inc/Baichuan2

Reverse engineering CAD models from raw geometry is a classic but strenuous research problem. Previous learning-based methods rely heavily on labels due to the supervised design patterns or reconstruct CAD shapes that are not easily editable. In this work, we introduce SECAD-Net, an end-to-end neural network aimed at reconstructing compact and easy-to-edit CAD models in a self-supervised manner. Drawing inspiration from the modeling language that is most commonly used in modern CAD software, we propose to learn 2D sketches and 3D extrusion parameters from raw shapes, from which a set of extrusion cylinders can be generated by extruding each sketch from a 2D plane into a 3D body. By incorporating the Boolean operation (i.e., union), these cylinders can be combined to closely approximate the target geometry. We advocate the use of implicit fields for sketch representation, which allows for creating CAD variations by interpolating latent codes in the sketch latent space. Extensive experiments on both ABC and Fusion 360 datasets demonstrate the effectiveness of our method, and show superiority over state-of-the-art alternatives including the closely related method for supervised CAD reconstruction. We further apply our approach to CAD editing and single-view CAD reconstruction. The code is released at https://github.com/BunnySoCrazy/SECAD-Net.

A long-standing goal of reinforcement learning is to acquire agents that can learn on training tasks and generalize well on unseen tasks that may share a similar dynamic but with different reward functions. The ability to generalize across tasks is important as it determines an agent's adaptability to real-world scenarios where reward mechanisms might vary. In this work, we first show that training a general world model can utilize similar structures in these tasks and help train more generalizable agents. Extending world models into the task generalization setting, we introduce a novel method named Task Aware Dreamer (TAD), which integrates reward-informed features to identify consistent latent characteristics across tasks. Within TAD, we compute the variational lower bound of sample data log-likelihood, which introduces a new term designed to differentiate tasks using their states, as the optimization objective of our reward-informed world models. To demonstrate the advantages of the reward-informed policy in TAD, we introduce a new metric called Task Distribution Relevance (TDR) which quantitatively measures the relevance of different tasks. For tasks exhibiting a high TDR, i.e., the tasks differ significantly, we illustrate that Markovian policies struggle to distinguish them, thus it is necessary to utilize reward-informed policies in TAD. Extensive experiments in both image-based and state-based tasks show that TAD can significantly improve the performance of handling different tasks simultaneously, especially for those with high TDR, and display a strong generalization ability to unseen tasks.

One-shot video-driven talking face generation aims at producing a synthetic talking video by transferring the facial motion from a video to an arbitrary portrait image. Head pose and facial expression are always entangled in facial motion and transferred simultaneously. However, the entanglement sets up a barrier for these methods to be used in video portrait editing directly, where it may require to modify the expression only while maintaining the pose unchanged. One challenge of decoupling pose and expression is the lack of paired data, such as the same pose but different expressions. Only a few methods attempt to tackle this challenge with the feat of 3D Morphable Models (3DMMs) for explicit disentanglement. But 3DMMs are not accurate enough to capture facial details due to the limited number of Blenshapes, which has side effects on motion transfer. In this paper, we introduce a novel self-supervised disentanglement framework to decouple pose and expression without 3DMMs and paired data, which consists of a motion editing module, a pose generator, and an expression generator. The editing module projects faces into a latent space where pose motion and expression motion can be disentangled, and the pose or expression transfer can be performed in the latent space conveniently via addition. The two generators render the modified latent codes to images, respectively. Moreover, to guarantee the disentanglement, we propose a bidirectional cyclic training strategy with well-designed constraints. Evaluations demonstrate our method can control pose or expression independently and be used for general video editing., Comment: https://carlyx.github.io/DPE/

Controlling the spin transport at the single-molecule level, especially without the use of ferromagnetic contacts, becomes a focus of research in spintronics. Inspired by the progress on atomic-level molecular synthesis, through first-principles calculations, we investigate the spin-dependent electronic transport of graphene nanoflakes with side-bonded functional groups, contacted by atomic carbon chain electrodes. It is found that, by rotating the functional group, the spin polarization of the transmission at the Fermi level could be switched between completely polarized and unpolarized states. Moreover, the transition between spin-up and spin-down polarized states can also be achieved, operating as a dual-spin filter. Further analysis shows that, it is the spin-dependent shift of density of states, caused by the rotation, that triggers the shift of transmission peaks, and then results in the variation of spin polarization. Such a feature is found to be robust to the of the nanoflake and the electrode material, showing great application potential. Those findings may throw light on the development of spintronic devices.

Model-based reinforcement learning usually suffers from a high sample complexity in training the world model, especially for the environments with complex dynamics. To make the training for general physical environments more efficient, we introduce Hamiltonian canonical ordinary differential equations into the learning process, which inspires a novel model of neural ordinary differential auto-encoder (NODA). NODA can model the physical world by nature and is flexible to impose Hamiltonian mechanics (e.g., the dimension of the physical equations) which can further accelerate training of the environment models. It can consequentially empower an RL agent with the robust extrapolation using a small amount of samples as well as the guarantee on the physical plausibility. Theoretically, we prove that NODA has uniform bounds for multi-step transition errors and value errors under certain conditions. Extensive experiments show that NODA can learn the environment dynamics effectively with a high sample efficiency, making it possible to facilitate reinforcement learning agents at the early stage.

Importance sampling (IS) is a popular technique in off-policy evaluation, which re-weights the return of trajectories in the replay buffer to boost sample efficiency. However, training with IS can be unstable and previous attempts to address this issue mainly focus on analyzing the variance of IS. In this paper, we reveal that the instability is also related to a new notion of Reuse Bias of IS -- the bias in off-policy evaluation caused by the reuse of the replay buffer for evaluation and optimization. We theoretically show that the off-policy evaluation and optimization of the current policy with the data from the replay buffer result in an overestimation of the objective, which may cause an erroneous gradient update and degenerate the performance. We further provide a high-probability upper bound of the Reuse Bias, and show that controlling one term of the upper bound can control the Reuse Bias by introducing the concept of stability for off-policy algorithms. Based on these analyses, we finally present a novel Bias-Regularized Importance Sampling (BIRIS) framework along with practical algorithms, which can alleviate the negative impact of the Reuse Bias. Experimental results show that our BIRIS-based methods can significantly improve the sample efficiency on a series of continuous control tasks in MuJoCo.

The buckling of spherical shells is plagued by a strong sensitivity to imperfections. Traditionally, imperfect shells tend to be characterized empirically by the knockdown factor, the ratio between the measured buckling strength and the corresponding classic prediction for a perfect shell. Recently, it has been demonstrated that the knockdown factor of a shell containing a single imperfection can be predicted when there is detailed a priori knowledge of the defect geometry. Still, addressing the analogous problem for a shell containing many defects remains an open question. Here, we use finite element simulations, which we validate against precision experiments, to investigate hemispherical shells containing a well-defined distribution of imperfections. Our goal is to characterize the resulting knockdown factor statistics. First, we study the buckling of shells containing only two defects, uncovering nontrivial regimes of interactions that echo existing findings for cylindrical shells. Then, we construct statistical ensembles of imperfect shells, whose defect amplitudes are sampled from a lognormal distribution. We find that a 3-parameter Weibull distribution is an excellent description for the measured statistics of knockdown factors, suggesting that shell buckling can be regarded as an extreme-value statistics phenomenon.

We investigate the mechanics of bistable, hard-magnetic, elastic beams, combining experiments, finite element modeling (FEM), and a reduced-order theory. The beam is made of a hard magneto-rheological elastomer, comprising two segments with antiparallel magnetization along the centerline, and is set into a bistable curved configuration by imposing an end-to-end shortening. Reversible snapping is possible between these two stable states. First, we experimentally characterize the critical field strength for the onset of snapping, at different levels of end-to-end shortening. Second, we perform 3D FEM simulations using the Riks method to analyze high-order deformation modes during snapping. Third, we develop a reduced-order centerline-based beam theory to rationalize the observed magneto-elastic response. The theory and simulations are validated against experiments, with an excellent quantitative agreement. Finally, we consider the case of combined magnetic and mechanical-indentation loading, examining how the applied field affects the bistability and quantifying the maximum load-bearing capacity. Our work provides a set of predictive tools for the rational design of one-dimensional, bistable, magneto-elastic structural elements.

Videos contain multi-modal content, and exploring multi-level cross-modal interactions with natural language queries can provide great prominence to text-video retrieval task (TVR). However, new trending methods applying large-scale pre-trained model CLIP for TVR do not focus on multi-modal cues in videos. Furthermore, the traditional methods simply concatenating multi-modal features do not exploit fine-grained cross-modal information in videos. In this paper, we propose a multi-level multi-modal hybrid fusion (M2HF) network to explore comprehensive interactions between text queries and each modality content in videos. Specifically, M2HF first utilizes visual features extracted by CLIP to early fuse with audio and motion features extracted from videos, obtaining audio-visual fusion features and motion-visual fusion features respectively. Multi-modal alignment problem is also considered in this process. Then, visual features, audio-visual fusion features, motion-visual fusion features, and texts extracted from videos establish cross-modal relationships with caption queries in a multi-level way. Finally, the retrieval outputs from all levels are late fused to obtain final text-video retrieval results. Our framework provides two kinds of training strategies, including an ensemble manner and an end-to-end manner. Moreover, a novel multi-modal balance loss function is proposed to balance the contributions of each modality for efficient end-to-end training. M2HF allows us to obtain state-of-the-art results on various benchmarks, eg, Rank@1 of 64.9\%, 68.2\%, 33.2\%, 57.1\%, 57.8\% on MSR-VTT, MSVD, LSMDC, DiDeMo, and ActivityNet, respectively., Comment: 1 1pages, 3 figures, 5 tables

We report on low-frequency current fluctuations, i.e. electronic noise, in quasi-one-dimensional (TaSe$_4$)$_2$I Weyl semimetal nanoribbons. It was found that the noise spectral density is of the 1/f type and scales with the square of the current, S~I^2 (f is the frequency). The noise spectral density increases by almost an order of magnitude and develops Lorentzian features near the temperature T~225 K. These spectral changes were attributed to the charge-density-wave phase transition even though the temperature of the noise maximum deviates from the reported Peierls transition temperature in bulk (TaSe$_4$)$_2$I crystals. The noise level, normalized by the channel area, in these Weyl semimetal nanoribbons was surprisingly low, $\sim 10^{-9}$ um$^2$Hz$^{-1}$ at f=10 Hz, when measured below and above the Peierls transition temperature. Obtained results shed light on the specifics of electron transport in quasi-1D topological Weyl semimetals and can be important for their proposed applications as downscaled interconnects., Comment: 20 pages, 5 figures

We propose a precise and efficient normal estimation method that can deal with noise and nonuniform density for unstructured 3D point clouds. Unlike existing approaches that directly take patches and ignore the local neighborhood relationships, which make them susceptible to challenging regions such as sharp edges, we propose to learn graph convolutional feature representation for normal estimation, which emphasizes more local neighborhood geometry and effectively encodes intrinsic relationships. Additionally, we design a novel adaptive module based on the attention mechanism to integrate point features with their neighboring features, hence further enhancing the robustness of the proposed normal estimator against point density variations. To make it more distinguishable, we introduce a multi-scale architecture in the graph block to learn richer geometric features. Our method outperforms competitors with the state-of-the-art accuracy on various benchmark datasets, and is quite robust against noise, outliers, as well as the density variations.

Registering urban point clouds is a quite challenging task due to the large-scale, noise and data incompleteness of LiDAR scanning data. In this paper, we propose SARNet, a novel semantic augmented registration network aimed at achieving efficient registration of urban point clouds at city scale. Different from previous methods that construct correspondences only in the point-level space, our approach fully exploits semantic features as assistance to improve registration accuracy. Specifically, we extract per-point semantic labels with advanced semantic segmentation networks and build a prior semantic part-to-part correspondence. Then we incorporate the semantic information into a learning-based registration pipeline, consisting of three core modules: a semantic-based farthest point sampling module to efficiently filter out outliers and dynamic objects; a semantic-augmented feature extraction module for learning more discriminative point descriptors; a semantic-refined transformation estimation module that utilizes prior semantic matching as a mask to refine point correspondences by reducing false matching for better convergence. We evaluate the proposed SARNet extensively by using real-world data from large regions of urban scenes and comparing it with alternative methods. The code is available at https://github.com/WinterCodeForEverything/SARNet., Comment: Author information changes

Though deep reinforcement learning (DRL) has obtained substantial success, it may encounter catastrophic failures due to the intrinsic uncertainty of both transition and observation. Most of the existing methods for safe reinforcement learning can only handle transition disturbance or observation disturbance since these two kinds of disturbance affect different parts of the agent; besides, the popular worst-case return may lead to overly pessimistic policies. To address these issues, we first theoretically prove that the performance degradation under transition disturbance and observation disturbance depends on a novel metric of Value Function Range (VFR), which corresponds to the gap in the value function between the best state and the worst state. Based on the analysis, we adopt conditional value-at-risk (CVaR) as an assessment of risk and propose a novel reinforcement learning algorithm of CVaR-Proximal-Policy-Optimization (CPPO) which formalizes the risk-sensitive constrained optimization problem by keeping its CVaR under a given threshold. Experimental results show that CPPO achieves a higher cumulative reward and is more robust against both observation and transition disturbances on a series of continuous control tasks in MuJoCo.

We develop a reduced-order model for thin plates made of hard magnetorheological elastomers (hard-MREs), which are materials composed of hard-magnetic particles embedded in a polymeric matrix. First, we propose a new magnetic potential, as an alternative to an existing torque-based 3D continuum theory of hard-MREs, obtained by reformulating the remnant magnetization of a deformed hard-MRE body. Specifically, the magnetizations in the initial and current configurations are related by the rotation tensor decomposed from the deformation gradient, independently of stretching deformation. This description is motivated by recently reported observations in microscopic homogenization simulations. Then, we derive a 2D plate model through the dimensional reduction of our proposed rotation-based 3D theory. For comparison, we also provide a second plate model derived from the existing 3D theory. Finally, we perform precision experiments to thoroughly evaluate the proposed 3D and 2D models on hard-magnetic plates under various magnetic and mechanical loading conditions. We demonstrate that our rotation-based modification of the magnetic potential is crucial in correctly capturing the behavior of plates subjected to an applied field aligned with the magnetization, and undergoing in-plane stretching. In all the tested cases, our rotation-based 3D and 2D models yield predictions in excellent quantitative agreement with the experiments and can thus serve as predictive tools for the rational design of hard-magnetic plate structures.

We study theoretically laser excitation of Rydberg triplet states of strontium atoms in the presence of weak microwave (MW) fields. Starting from the ground state $5s^2\,^1S_0$, the Rydberg excitation is realized through the metastable, triplet $5s5p\,^3P_1$ state, whose decay rate $\gamma_2$ is $2\pi\times 7.5$ kHz, much smaller than the one in the singlet state or alkali atoms. The influences of $\gamma_2$ on the transparency and absorption spectrum in the electromagnetically induced transparency (EIT), and electromagnetically induced absorption (EIA) regime are examined. Narrow transparent windows (EIT) or absorption peaks (EIA) are found, whose distance in the spectrum depends on the Rabi frequency of the weak MW field. It is found that the spectrum exhibits higher contrast than using the singlet state or alkali atoms in typical situations. Using the metastable intermediate state, we find that resonance fluorescence of Sr gases exhibits very narrow peaks, which are modulated by the MW field. When the MW field is weaker than the probe and control light, the spectrum distance of these peaks are linearly proportional to $\Omega_m$. This allows us to propose a new way to sense very weak MW fields through resonance fluorescence. Our study shows that the Sr triplet state could be used to develop the Rydberg MW electrometry that gains unique advantages.

In high-stake scenarios like medical treatment and auto-piloting, it's risky or even infeasible to collect online experimental data to train the agent. Simulation-based training can alleviate this issue, but may suffer from its inherent mismatches from the simulator and real environment. It is therefore imperative to utilize the simulator to learn a robust policy for the real-world deployment. In this work, we consider policy learning for Robust Markov Decision Processes (RMDP), where the agent tries to seek a robust policy with respect to unexpected perturbations on the environments. Specifically, we focus on the setting where the training environment can be characterized as a generative model and a constrained perturbation can be added to the model during testing. Our goal is to identify a near-optimal robust policy for the perturbed testing environment, which introduces additional technical difficulties as we need to simultaneously estimate the training environment uncertainty from samples and find the worst-case perturbation for testing. To solve this issue, we propose a generic method which formalizes the perturbation as an opponent to obtain a two-player zero-sum game, and further show that the Nash Equilibrium corresponds to the robust policy. We prove that, with a polynomial number of samples from the generative model, our algorithm can find a near-optimal robust policy with a high probability. Our method is able to deal with general perturbations under some mild assumptions and can also be extended to more complex problems like robust partial observable Markov decision process, thanks to the game-theoretical formulation., Comment: AAAI 2022

The relationship between charge-density-wave (CDW) and superconductivity (SC), two vital physical phases in condensed matter physics, has always been the focus of scientists' research over the past decades. Motivated by this research hotspot, we systematically studied the physical properties of the layered telluride chalcogenide superconductors CuIr2-xAlxTe4. Through the resistance and magnetization measurements, we found that the CDW order was destroyed by a small amount of Al doping. Meanwhile, the superconducting transition temperature (Tc) kept changing with the change of doping amount and rose towards the maximum value of 2.75 K when x = 0.075. The value of normalized specific heat jump for the highest Tc sample CuIr1.925Al0.075Te4 was 1.53, which was larger than the BCS value of 1.43 and showed that bulk superconducting nature. In order to clearly show the relationship between SC and CDW states, we propose a phase diagram of Tc vs. doping content., Comment: 13 pages, 6 figures, 2 tables; The manuscript with the same title will be published by Chinese Physics B

Reconstructing high-fidelity 3D facial texture from a single image is a quite challenging task due to the lack of complete face information and the domain gap between the 3D face and 2D image. Further, obtaining re-renderable 3D faces has become a strongly desired property in many applications, where the term 're-renderable' demands the facial texture to be spatially complete and disentangled with environmental illumination. In this paper, we propose a new self-supervised deep learning framework for reconstructing high-quality and re-renderable facial albedos from single-view images in-the-wild. Our main idea is to first utilize a prior generation module based on the 3DMM proxy model to produce an unwrapped texture and a globally parameterized prior albedo. Then we apply a detail refinement module to synthesize the final texture with both high-frequency details and completeness. To further make facial textures disentangled with illumination, we propose a novel detailed illumination representation which is reconstructed with the detailed albedo together. We also design several novel regularization losses on both the albedo and illumination maps to facilitate the disentanglement of these two factors. Finally, by leveraging a differentiable renderer, each face attribute can be jointly trained in a self-supervised manner without requiring ground-truth facial reflectance. Extensive comparisons and ablation studies on challenging datasets demonstrate that our framework outperforms state-of-the-art approaches.

We develop a reduced model for hard-magnetic, thin, linear-elastic shells that can be actuated through an external magnetic field, with geometrically exact strain measures. Assuming a reduced kinematics based on the Kirchhoff-Love assumption, we derive a reduced two-dimensional magneto-elastic energy that can be minimized through numerical analysis. In parallel, we simplify the reduced energy by expanding it up to the second-order in the displacement field and provide a physical interpretation. Our theoretical analysis allows us to identify and interpret the two primary mechanisms dictating the magneto-elastic response: a combination of equivalent magnetic pressure and forces at the first order, and distributed magnetic torques at the second order. We contrast our reduced framework against a three-dimensional nonlinear model by investigating three test cases involving the indentation and the pressure buckling of shells under magnetic loading. We find excellent agreement between the two approaches, thereby verifying our reduced model for shells undergoing nonlinear and non-axisymmetric deformations. We believe that our model for magneto-elastic shells will serve as a valuable tool for the rational design of magnetic structures, enriching the set of reduced magnetic models., Comment: 24 pages, 5 figures

Point set registration is an essential step in many computer vision applications, such as 3D reconstruction and SLAM. Although there exist many registration algorithms for different purposes, however, this topic is still challenging due to the increasing complexity of various real-world scenarios, such as heavy noise and outlier contamination. In this paper, we propose a novel probabilistic generative method to simultaneously align multiple point sets based on the heavy-tailed Laplacian distribution. The proposed method assumes each data point is generated by a Laplacian Mixture Model (LMM), where its centers are determined by the corresponding points in other point sets. Different from the previous Gaussian Mixture Model (GMM) based method, which minimizes the quadratic distance between points and centers of Gaussian probability density, LMM minimizes the sparsity-induced L1 distance, thereby it is more robust against noise and outliers. We adopt Expectation-Maximization (EM) framework to solve LMM parameters and rigid transformations. We approximate the L1 optimization as a linear programming problem by exponential mapping in Lie algebra, which can be effectively solved through the interior point method. To improve efficiency, we also solve the L1 optimization by Alternating Direction Multiplier Method (ADMM). We demonstrate the advantages of our method by comparing it with representative state-of-the-art approaches on benchmark challenging data sets, in terms of robustness and accuracy.

Lensless cameras are a class of imaging devices that shrink the physical dimensions to the very close vicinity of the image sensor by replacing conventional compound lenses with integrated flat optics and computational algorithms. Here we report a diffractive lensless camera with spatially-coded Voronoi-Fresnel phase to achieve superior image quality. We propose a design principle of maximizing the acquired information in optics to facilitate the computational reconstruction. By introducing an easy-to-optimize Fourier domain metric, Modulation Transfer Function volume (MTFv), which is related to the Strehl ratio, we devise an optimization framework to guide the optimization of the diffractive optical element. The resulting Voronoi-Fresnel phase features an irregular array of quasi-Centroidal Voronoi cells containing a base first-order Fresnel phase function. We demonstrate and verify the imaging performance for photography applications with a prototype Voronoi-Fresnel lensless camera on a 1.6-megapixel image sensor in various illumination conditions. Results show that the proposed design outperforms existing lensless cameras, and could benefit the development of compact imaging systems that work in extreme physical conditions., Comment: 17 pages, 11 figures, 1 video, 1 video description, and 1 supplemental document

Removing undesirable specular highlight from a single input image is of crucial importance to many computer vision and graphics tasks. Existing methods typically remove specular highlight for medical images and specific-object images, however, they cannot handle the images with text. In addition, the impact of specular highlight on text recognition is rarely studied by text detection and recognition community. Therefore, in this paper, we first raise and study the text-aware single image specular highlight removal problem. The core goal is to improve the accuracy of text detection and recognition by removing the highlight from text images. To tackle this challenging problem, we first collect three high-quality datasets with fine-grained annotations, which will be appropriately released to facilitate the relevant research. Then, we design a novel two-stage network, which contains a highlight detection network and a highlight removal network. The output of highlight detection network provides additional information about highlight regions to guide the subsequent highlight removal network. Moreover, we suggest a measurement set including the end-to-end text detection and recognition evaluation and auxiliary visual quality evaluation. Extensive experiments on our collected datasets demonstrate the superior performance of the proposed method.

In this paper, we present Tianshou, a highly modularized Python library for deep reinforcement learning (DRL) that uses PyTorch as its backend. Tianshou intends to be research-friendly by providing a flexible and reliable infrastructure of DRL algorithms. It supports online and offline training with more than 20 classic algorithms through a unified interface. To facilitate related research and prove Tianshou's reliability, we have released Tianshou's benchmark of MuJoCo environments, covering eight classic algorithms with state-of-the-art performance. We open-sourced Tianshou at https://github.com/thu-ml/tianshou/.

Thin beams made of magnetorheological elastomers embedded with hard magnetic particles (hard-MREs) are capable of large deflections under an applied magnetic field. We propose a comprehensive framework, comprising a beam model and 3D finite element modeling (FEM), to describe the behavior of hard-MRE beams under both uniform and constant gradient magnetic fields. First, based on the Helmholtz free energy of bulk (3D) hard-MREs, we perform dimensional reduction to derive a 1D description and obtain the equilibrium equation of the beam through variational methods. In parallel, we extend the existing 3D continuum theory for hard-MREs to the general case of non-uniform fields by incorporating the magnetic body force induced by the field gradient and implementing it in FEM. The beam model and FEM are first validated using experiments and then employed to predict the deflection of a cantilever beam in either a uniform or a constant gradient field. The corresponding parameters governing the magneto-elastic coupling are identified. Then, a set of comparative numerical studies for actuation in different configurations yields additional insight into the beam response. Our study builds on previous work on hard-MRE beams, while providing a more complete framework, both in terms of the methodologies used and the configurations considered, to serve as a valuable predictive toolbox for the rational design of beam-like hard-magnetic structures.

In this paper, we study the graph of homothety classes of stable free lattices in a two-dimensional representation over a local UFD. This generalizes a classical result of the case where the base ring is a discrete valuation ring due to Serre. As applications, we consider the case when the representation comes from a residually reducible Hida family and we study the control theorem of Selmer groups. These results enable us to know the precise statement of the main conjecture in residually reducible case as we will remark in section 4 and section 5., Comment: 30 pages, comments welcome

Besides magnetic and charge order, regular arrangements of orbital occupation constitute a fundamental order parameter of condensed matter physics. Even though orbital order is difficult to identify directly in experiments, its presence was firmly established in a number of strongly correlated, three-dimensional Mott insulators. Here, reporting resonant X-ray scattering experiments on the layered Van der Waals compound $1T$-TiSe$_2$, we establish the emergence of orbital order in a weakly correlated, quasi-two-dimensional material. Our experimental scattering results are consistent with first-principles calculations that bring to the fore a generic mechanism of close interplay between charge redistribution, lattice displacements, and orbital order. It demonstrates the essential role that orbital degrees of freedom play in TiSe$_2$, and their importance throughout the family of correlated Van der Waals materials.

Face recognition is an important yet challenging problem in computer vision. A major challenge in practical face recognition applications lies in significant variations between profile and frontal faces. Traditional techniques address this challenge either by synthesizing frontal faces or by pose invariant learning. In this paper, we propose a novel method with Lie algebra theory to explore how face rotation in the 3D space affects the deep feature generation process of convolutional neural networks (CNNs). We prove that face rotation in the image space is equivalent to an additive residual component in the feature space of CNNs, which is determined solely by the rotation. Based on this theoretical finding, we further design a Lie Algebraic Residual Network (LARNet) for tackling pose robust face recognition. Our LARNet consists of a residual subnet for decoding rotation information from input face images, and a gating subnet to learn rotation magnitude for controlling the strength of the residual component contributing to the feature learning process. Comprehensive experimental evaluations on both frontal-profile face datasets and general face recognition datasets convincingly demonstrate that our method consistently outperforms the state-of-the-art ones., Comment: Accepted by ICML 2021

Shell buckling is central in many biological structures and advanced functional materials, even if, traditionally, this elastic instability has been regarded as a catastrophic phenomenon to be avoided for engineering structures. Either way, predicting critical buckling conditions remains a long-standing challenge. The subcritical nature of shell buckling imparts extreme sensitivity to material and geometric imperfections. Consequently, measured critical loads are inevitably lower than classic theoretical predictions. Here, we present a robust mechanism to dynamically tune the buckling strength of shells, exploiting the coupling between mechanics and magnetism. Our experiments on pressurized spherical shells made of a magnetorheological elastomer demonstrate the tunability of their buckling pressure via magnetic actuation. We develop a theoretical model for thin magnetic elastic shells, which rationalizes the underlying mechanism, in excellent agreement with experiments. A dimensionless magneto-elastic buckling number is recognized as the key governing parameter, combining the geometrical, mechanical, and magnetic properties of the system.

Recent learning-based approaches show promising performance improvement for scene text removal task. However, these methods usually leave some remnants of text and obtain visually unpleasant results. In this work, we propose a novel "end-to-end" framework based on accurate text stroke detection. Specifically, we decouple the text removal problem into text stroke detection and stroke removal. We design a text stroke detection network and a text removal generation network to solve these two sub-problems separately. Then, we combine these two networks as a processing unit, and cascade this unit to obtain the final model for text removal. Experimental results demonstrate that the proposed method significantly outperforms the state-of-the-art approaches for locating and erasing scene text. Since current publicly available datasets are all synthetic and cannot properly measure the performance of different methods, we therefore construct a new real-world dataset, which will be released to facilitate the relevant research., Comment: 14 pages, 9 figures

Recent GAN-based image inpainting approaches adopt an average strategy to discriminate the generated image and output a scalar, which inevitably lose the position information of visual artifacts. Moreover, the adversarial loss and reconstruction loss (e.g., l1 loss) are combined with tradeoff weights, which are also difficult to tune. In this paper, we propose a novel detection-based generative framework for image inpainting, which adopts the min-max strategy in an adversarial process. The generator follows an encoder-decoder architecture to fill the missing regions, and the detector using weakly supervised learning localizes the position of artifacts in a pixel-wise manner. Such position information makes the generator pay attention to artifacts and further enhance them. More importantly, we explicitly insert the output of the detector into the reconstruction loss with a weighting criterion, which balances the weight of the adversarial loss and reconstruction loss automatically rather than manual operation. Experiments on multiple public datasets show the superior performance of the proposed framework. The source code is available at https://github.com/Evergrow/GDN_Inpainting., Comment: 12 pages, 9 figures, accepted by Computer Graphics Forum, supplementary material link: https://evergrow.github.io/GDN_Inpainting_files/GDN_Inpainting_Supplement.pdf

The integration of satellite and terrestrial communication systems plays a vital role in the fifth-generation mobile communication system (5G) for the ubiquitous coverage, reliable service and flexible networking. Moreover, the millimeter wave (mmWave) communication with large bandwidth is a key enabler for 5G intelligent rail transportation. In this paper, the satellite-terrestrial channel at 22.6 GHz is characterized for a typical high-speed railway (HSR) environment. The three-dimensional model of the railway scenario is reconstructed and imported into the Cloud Ray-Tracing (CloudRT) simulation platform. Based on extensive ray-tracing simulations, the channel for the terrestrial HSR system and the satellite-terrestrial system with two weather conditions are characterized, and the interference between them are evaluated. The results of this paper can help for the design and evaluation for the satellite-terrestrial communication system enabling future intelligent rail transportation.

The interplay between superconductivity and charge density wave (CDW)/metal-to-insulator transition (MIT) has long been interested and studied in condensed matter physics. Here we study systematically the charge density wave and superconductivity properties in the solid solutions Zn1-xCuxIr2-yN(N = Al, Ti, Rh)yTe4. Resistivity, magnetic susceptibility and specific heat measurements indicate that the CDW state was suppressed immediately while the superconducting critical temperature (Tc) differs from each system. In the Al- and Ti-substitution cases, Tc increase as y increases and reaches a maximum around 2.75 K and 2.84 K respectively at y = 0.075, followed by a decrease of Tc before the chemical phase boundary is reached at y = 0.2. Nevertheless, Tc decreases monotonously with Rh-doping content y increases and disappears above 0.3 with measuring temperature down to 2 K. Surprisingly, in the Zn1-xCuxIr2Te4 solid solution, Tc enhances as x increases and reaches a maximum value of 2.82 K for x = 0.5 but subsequently survives over the whole doping range of 0.00 - 0.9 despite Tc changes slightly with higher doping content, which differs from the observation of zinc doping suppressing the superconductivity quickly in the high Tc cuprate superconductors. The specific heat anomaly at the superconducting transitions for the representative optimal doping samples are all slightly higher than the BCS value of 1.43 and indicate bulk superconductivity in these compounds. Finally, the CDW transition temperature (TCDW) and superconducting transition temperature (Tc) vs. x/y content phase diagrams of Zn1-xCuxIr2-yN(N = Al, Ti, Rh)yTe4 have been established and compared, which offers good opportunity to study the competition between CDW and superconductivity in the telluride chalcogenides., Comment: 25 pages, 10 figures, 1 table. arXiv admin note: text overlap with arXiv:1908.09292

The interactions between players of prisoner's dilemma (PD) game are reconstructed with evolutionary game data. All participants play the game with their counterparts and gain corresponding rewards during each round of the game. However, their strategies are updated asynchronously during the evolutionary PD game. Two inference methods of the interactions between players are derived with naive mean-field (nMF) approximation and maximum log-likelihood estimation (MLE) respectively. The two methods are tested numerically also for fully connected asymmetric Sherrington-Kirkpatrick (SK) models, varying the data length, asymmetric degree, payoff and system noise (coupling strength). We find that the reconstruction mean square error (MSE) of MLE method is proportional to the inverse of data length and typically half (benefit from the extra information of update times) of that by nMF. Both methods are robust to the asymmetric degree but works better for large payoff. Compared with MLE, nMF is more sensitive to the couplings strength which prefers weak couplings.

We propose a novel framework for computing descriptors for characterizing points on three-dimensional surfaces. First, we present a new non-learned feature that uses graph wavelets to decompose the Dirichlet energy on a surface. We call this new feature wavelet energy decomposition signature (WEDS). Second, we propose a new multiscale graph convolutional network (MGCN) to transform a non-learned feature to a more discriminative descriptor. Our results show that the new descriptor WEDS is more discriminative than the current state-of-the-art non-learned descriptors and that the combination of WEDS and MGCN is better than the state-of-the-art learned descriptors. An important design criterion for our descriptor is the robustness to different surface discretizations including triangulations with varying numbers of vertices. Our results demonstrate that previous graph convolutional networks significantly overfit to a particular resolution or even a particular triangulation, but MGCN generalizes well to different surface discretizations. In addition, MGCN is compatible with previous descriptors and it can also be used to improve the performance of other descriptors, such as the heat kernel signature, the wave kernel signature, or the local point signature., Comment: Accepted to SIGGRAPH 2020. (15 pages, 15 figures, 12 tables, low-resolution version)

Here we report the effect of structural and superconductivity properties on Ru doped CuIr2Te4 telluride chalcogenide. XRD results suggest that the CuIr2-xRuxTe4 maintain the disordered trigonal structure with space group P3m1 (No. 164) for x less than 0.3. The lattice constants, a and c, both decrease with increasing Ru content. Temperature-dependent resistivity, magnetic susceptibility and specific-heat measurements are performed to characterize the superconducting properties systematically. Our results suggest that the optimal doping level for superconductivity in CuIr2-xRuxTe4 is x = 0.05, where Tc is 2.79 K with the Sommerfeld constant gamma of 11.52 mJ mol-1 K-2 and the specific-heat anomaly at the superconducting transition, is approximately 1.51, which is higher than the BCS value of 1.43, indicating CuIr1.95Ru0.05Te4 is a strongly electron-phonon coupled superconductor. The values of lower critical filed and upper critical field calculated from isothermal magnetization and magneto-transport measurements are 0.98 KOe and 2.47 KOe respectively, signifying that the compound is clearly a type-II superconductor. Finally, a dome-like shape superconducting Tcs vs. x content phase diagram is established, where the charge density wave disappears at x = 0.03 while superconducting transition temperature (Tc) rises until it reaches its peak at x = 0.05, then, with decreasing when x reaches 0.3. This feature of the competition between CDW and the superconductivity could be caused by tuning the Fermi surface and density of states with Ru chemical doping., Comment: 19 pages, 6 figures, 2 tables. arXiv admin note: text overlap with arXiv:1908.05438

Here we report the first observation of superconductivity in the AB2X4-type ternary telluride CuIr2Te4, which is synthesized by a solid-state method in an evacuated quartz jacket. It adopts a disordered trigonal structure with space group P3m1 (No. 164), which embodies a two-dimensional IrTe2 layers and intercalated by Cu between the layers. We use a combination of experimental and first principles calculation analysis to look insight into the structural and physical properties. CuIr2Te4 consistently exhibited a bulk superconductivity transition at 2.5 K in electrical resistivity, magnetic susceptibility and specific heat measurements. Resistivity and magnetization measurements suggest a charge density wave transition (TCDW = 250 K) coexists in CuIr2Te4, which further signify the coexistence of superconductivity and CDW in the ternary telluride chalcogenide CuIr2Te4. Our discovery of the new CDW-bearing superconductor CuIr2Te4 opens a door for experimental and theoretical studies of the interplay between CDW and superconductivity quantum state in the condensed matter field., Comment: 25 pages, 6 figures, 2 tables

High stability and oxygen permeability are two prominent requirements for the oxygen transport membrane candidates used as industrialization. Herein, we report several composite membranes based on xwt.%Ce0.9Pr0.1O2(CPO)-(100-x)wt.%Pr0.6Sr0.4Fe0.8Al0.2O3(PSFAO) (x = 50, 60 and 75) prepared via a modified Pechini method. Oxygen permeability test reveals that the 60CPO-40PSFAO composition exhibits the highest oxygen permeability. The oxygen permeation flux through the optimal uncoated 0.33 mm-thickness 60CPO-40PSFAO composite can reach 1.03 mL cm-2 min-1 (over the general requirement value of 1 mL cm-2 min-1) in air/He atmosphere at 1000 {\deg}C. In situ XRD performance confirms the optimal 60CPO-40PSFAO sample shows excellent stability in CO2-containing atmospheres. The 60CPO-40PSFAO membrane still exhibits simultaneously excellent oxygen permeability and phase stability after operating for over 100 h at air/CO2 condition at 1000 {\deg}C, which further indicates that the 60CPO-40PSFAO composite is likely to be used for oxygen supply in CO2 capture, Comment: 29 pages,12 figures, 1 table

Transition metal dichalcogenides (TMDCs) usually exhibit layered polytypic structures due to the weak interlayer coupling. 2H-NbSe2 is one of the most widely studied in the pristine TMDC family due to its high superconducting transition temperature (Tc = 7.3K) and the occurrence of a charge-density wave (CDW) order below 33 K. The coexistence of CDW with superconductivity poses an intriguing open question about the relationship between Fermi surface nesting and Cooper pairing. Past studies of this issue have mostly been focused on doping 2H-NbSe2 by 3d transition metals without significantly changing its crystal structure. Here we replaced the Se by Te in 2H-NbSe2 in order to design a new 1T polytype layered TMDC NbSeTe, which adopts a trigonal structure with space group P-3m1. We successfully grew large size and high-quality single crystals of 1T-NbSeTe via the vapor transport method using I2 as the transport agent. Temperature-dependent resistivity and specific heat data revealed a bulk Tc at 1.3 K, which is the first observation of superconductivity in pure 1T-NbSeTe phase. This compound enlarged the family of superconducting TMDCs and provides an opportunity to study the interplay between CDW and superconductivity in the trigonal structure., Comment: 21 pages, 6 figures, 2 tables

2H-NbSe2 is one of the most widely researched transition metal dichalcogenide (TMD) superconductors, which undergoes charge-density wave (CDW) transition at TCDW about 33 K and superconducting transition at Tc of 7.3 K. To explore the relation between its superconductivity and Fermi surface nesting, we combined S substitution with Cu intercalation in 2H-NbSe2 to make CuxNbSe2-ySy. Upon systematic substitution of S and intercalation of Cu ions into 2H-NbSe2, we found that when the Cu and S contents increases, the Tc decreases in CuxNbSe2-ySy. While at higher x and y values, Tc keeps a constant value near 2 K, which is not commonly observed for a layered TMD. For comparison, we found the simultaneous substitution of Nb by Cu and Se by S in CuxNb1-xSe2-ySy lowered the Tc substantially faster. We construct a superconducting phase diagrams for our double-doping compounds in contrast with the related single-ions doping systems., Comment: 20 pages, 6 figures, post-print of the article published in the Superconductor Science and Technology (2019)