arXiv subjects

Wenxuan Zhu

Publications and source records attributed to Wenxuan Zhu.

31 records · Page 2Linked to original sources

arXiv2024

Vivid-ZOO: Multi-View Video Generation with Diffusion Model

While diffusion models have shown impressive performance in 2D image/video generation, diffusion-based Text-to-Multi-view-Video (T2MVid) generation remains underexplored. The new challenges posed by T2MVid generation lie in the lack of massive captioned multi-view videos and the complexity of modeling such multi-dimensional distribution. To this end, we propose a novel diffusion-based pipeline that generates high-quality multi-view videos centered around a dynamic 3D object from text. Specifically, we factor the T2MVid problem into viewpoint-space and time components. Such factorization allows us to combine and reuse layers of advanced pre-trained multi-view image and 2D video diffusion models to ensure multi-view consistency as well as temporal coherence for the generated multi-view videos, largely reducing the training cost. We further introduce alignment modules to align the latent spaces of layers from the pre-trained multi-view and the 2D video diffusion models, addressing the reused layers' incompatibility that arises from the domain gap between 2D and multi-view data. In support of this and future research, we further contribute a captioned multi-view video dataset. Experimental results demonstrate that our method generates high-quality multi-view videos, exhibiting vivid motions, temporal coherence, and multi-view consistency, given a variety of text prompts.

cs.CV↗

arXiv2024

Exploring Learngene via Stage-wise Weight Sharing for Initializing Variable-sized Models

In practice, we usually need to build variable-sized models adapting for diverse resource constraints in different application scenarios, where weight initialization is an important step prior to training. The Learngene framework, introduced recently, firstly learns one compact part termed as learngene from a large well-trained model, after which learngene is expanded to initialize variable-sized models. In this paper, we start from analysing the importance of guidance for the expansion of well-trained learngene layers, inspiring the design of a simple but highly effective Learngene approach termed SWS (Stage-wise Weight Sharing), where both learngene layers and their learning process critically contribute to providing knowledge and guidance for initializing models at varying scales. Specifically, to learn learngene layers, we build an auxiliary model comprising multiple stages where the layer weights in each stage are shared, after which we train it through distillation. Subsequently, we expand these learngene layers containing stage information at their corresponding stage to initialize models of variable depths. Extensive experiments on ImageNet-1K demonstrate that SWS achieves consistent better performance compared to many models trained from scratch, while reducing around 6.6x total training costs. In some cases, SWS performs better only after 1 epoch tuning. When initializing variable-sized models adapting for different resource constraints, SWS achieves better results while reducing around 20x parameters stored to initialize these models and around 10x pre-training costs, in contrast to the pre-training and fine-tuning approach.

cs.LG↗

arXiv2024

Electrical-controllable antiferromagnet-based tunnel junction

Electrical-controllable antiferromagnet tunnel junction is a key goal in spintronics, holding immense promise for ultra-dense and ultra-stable antiferromagnetic memory with high processing speed for modern information technology. Here, we have advanced towards this goal by achieving an electrical-controllable antiferromagnet-based tunnel junction of Pt/Co/Pt/Co/IrMn/MgO/Pt. The exchange coupling between antiferromagnetic IrMn and Co/Pt perpendicular magnetic multilayers results in the formation of interfacial exchange bias and exchange spring in IrMn. Encoding information states 0 and 1 is realized through the exchange spring in IrMn, which can be electrically written by spin-orbit torque switching with high cyclability and electrically read by antiferromagnetic tunneling anisotropic magnetoresistance. Combining spin-orbit torque switching of both exchange spring andexchange bias, 16 Boolean logic operation is successfully demonstrated. With both memory and logic functionalities integrated into our electrical-controllable antiferromagnetic-based tunnel junction, we chart the course toward high-performance antiferromagnetic logic-in-memory.

physics.app-ph↗

arXiv2024

Observation of non-volatile anomalous Nernst effect in altermagnet with collinear Néel vector

Anomalous Nernst effect (ANE), a widely investigated transverse thermoelectric effect that converts waste heat into electrical energy with remarkable flexibility and integration capability, has been extended to antiferromagnets with non-collinear spin texture recently. ANE in compensated magnet with collinear Néel vector will bring more opportunities to construct magnetic-field-immune and ultrafast transverse thermoelectric converters, but remains unachieved for long. It is due to the degenerated band structure of traditional collinear compensated magnet excludes non-zero Berry curvature. Here, we realize non-volatile ANE in altermagnet Mn5Si3 thin film with collinear Neel vector, whose unique alternating spin-splitting band structure plays vital role in creating non-zero Berry curvature and hotpots of anomalous Nernst conductivity near band intersections. Interestingly, ANE is relatively weak in stoichiometric Mn5Si3, but undergoes a sixfold enhancement through strategically raising the Fermi level by additional Mn doping, indicating sensitive intrinsic influence from specific location of the Fermi level on ANE in altermagnet. Moreover, our investigation reveals a unique Neel-vector-dependent temperature-scaling relationship of anomalous Nernst conductivity in Mn5Si3. Our work not only fills a longstanding gap by confirming the presence of non-volatile ANE in collinear compensated magnet, but also enlightens thermoelectric physics related to exotic spin-splitting band structure in altermagnet.

cond-mat.mtrl-sci↗

arXiv2024

Electrical 180o switching of Néel vector in spin-splitting antiferromagnet

Antiferromagnetic spintronics have attracted wide attention due to its great potential in constructing ultra-dense and ultra-fast antiferromagnetic memory that suits modern high-performance information technology. The electrical 180o switching of Néel vector is a long-term goal for developing electrical-controllable antiferromagnetic memory with opposite Néel vectors as binary "0" and "1". However, the state-of-art antiferromagnetic switching mechanisms have long been limited for 90o or 120o switching of Néel vector, which unavoidably require multiple writing channels that contradicts ultra-dense integration. Here, we propose a deterministic switching mechanism based on spin-orbit torque with asymmetric energy barrier, and experimentally achieve electrical 180o switching of spin-splitting antiferromagnet Mn5Si3. Such a 180o switching is read out by the Néel vector-induced anomalous Hall effect. Based on our writing and readout methods, we fabricate an antiferromagnet device with electrical-controllable high and low resistance states that accomplishes robust write and read cycles. Besides fundamental advance, our work promotes practical spin-splitting antiferromagnetic devices based on spin-splitting antiferromagnet.

cond-mat.mtrl-sci↗

arXiv2023

Causal Deep Reinforcement Learning Using Observational Data

Deep reinforcement learning (DRL) requires the collection of interventional data, which is sometimes expensive and even unethical in the real world, such as in the autonomous driving and the medical field. Offline reinforcement learning promises to alleviate this issue by exploiting the vast amount of observational data available in the real world. However, observational data may mislead the learning agent to undesirable outcomes if the behavior policy that generates the data depends on unobserved random variables (i.e., confounders). In this paper, we propose two deconfounding methods in DRL to address this problem. The methods first calculate the importance degree of different samples based on the causal inference technique, and then adjust the impact of different samples on the loss function by reweighting or resampling the offline dataset to ensure its unbiasedness. These deconfounding methods can be flexibly combined with existing model-free DRL algorithms such as soft actor-critic and deep Q-learning, provided that a weak condition can be satisfied by the loss functions of these algorithms. We prove the effectiveness of our deconfounding methods and validate them experimentally.

cs.LG↗

arXiv2023

Edge-aware Plug-and-play Scheme for Semantic Segmentation

Semantic segmentation is a classic and fundamental computer vision problem dedicated to assigning each pixel with its corresponding class. Some recent methods introduce edge-based information for improving the segmentation performance. However these methods are specific and limited to certain network architectures, and they can not be transferred to other models or tasks. Therefore, we propose an abstract and universal edge supervision method called Edge-aware Plug-and-play Scheme (EPS), which can be easily and quickly applied to any semantic segmentation models. The core is edge-width/thickness preserving guided for semantic segmentation. The EPS first extracts the Edge Ground Truth (Edge GT) with a predefined edge thickness from the training data; and then for any network architecture, it directly copies the decoder head for the auxiliary task with the Edge GT supervision. To ensure the edge thickness preserving consistantly, we design a new boundarybased loss, called Polar Hausdorff (PH) Loss, for the auxiliary supervision. We verify the effectiveness of our EPS on the Cityscapes dataset using 22 models. The experimental results indicate that the proposed method can be seamlessly integrated into any state-of-the-art (SOTA) models with zero modification, resulting in promising enhancement of the segmentation performance.

cs.CV↗

arXiv2022

Ordered creation and motion of skyrmions with surface acoustic wave

Magnetic skyrmions with a well-defined spin texture have shown unprecedented potential for various spintronic applications owning to their topologically non-trivial and quasiparticle properties. To put skyrmions into practical technology, efficient manipulation, especially the inhibition of skyrmion Hall effect (SkHE) has been intensively pursued. In spite of the recent progress made on reducing SkHE in several substituted systems, such as ferrimagnets and synthetic antiferromagnets, the organized creation and current driven motion of skyrmions with negligible SkHE in ferromagnets remain challenging. Here, by embeding the [Co/Pd] multilayer into a surface acoustic wave (SAW) delay line, we experimentally realized the ordered generation of magnetic skyrmions. The resultant current-induced skyrmions movement with negligible SkHE was observed, which can be attributed to the energy redistribution of the system during the excitation of SAW. Our findings open up an unprecedentedly new perspective for manipulating topological solitons, and would advance the development of skyrmionics and spin acousto-electronics.

cond-mat.mtrl-sci↗

arXiv2022

Observation of spin splitting torque in a collinear antiferromagnet RuO2

Current-induced spin torques provide efficient data writing approaches for magnetic memories. Recently, the spin splitting torque (SST) was theoretically predicted (R. González-Hernández et al. Phys. Rev. Lett. 126, 127701 (2021)), which combines advantages of conventional spin transfer torque (STT) and spin-orbit torque (SOT) as well as enables controllable spin polarization. Here we provide the experimental evidence of SST in collinear antiferromagnet RuO2 films. The spin current direction is found to be correlated to the crystal orientation of RuO2 and the spin polarization direction is dependent on (parallel to) the Néel vector. These features are quite characteristic for the predicted SST. Our finding not only present a new member for the spin torques besides traditional STT and SOT, but also proposes a promising spin source RuO2 for spintronics.

cond-mat.mtrl-sci↗

arXiv2022

Controllable anomalous Nernst effect in an antiperovskite antiferromagnet

Anomalous Nernst effect (ANE), the generation of a transverse electric voltage by a longitudinal temperature gradient, has attracted increasing interests of researchers recently, due to its potential in the thermoelectric power conversion and close relevance to the Berry curvature of the band structure. Avoiding the stray field of ferromagnets, ANE in antiferromagnets (AFM) has the advantage of realizing highly efficient and densely integrated thermopiles. Here, we report the observation of ANE in an antiperovskite noncollinear AFM Mn3SnN experimentally, which is triggered by the enhanced Berry curvature from Weyl points located close to the Fermi level. Considering that antiperovskite Mn3SnN has rich magnetic phase transition, we modulate the noncollinear AFM configurations by the biaxial strain, which enables us to control its ANE. Our findings provide a potential class of materials to explore the Weyl physics of noncollinear AFM as well as realizing antiferromagnetic spin caloritronics that exhibits promising prospects for energy conversion and information processing.

cond-mat.mtrl-sci↗

arXiv2021

Noble-Metal-Assisted Fast Interfacial Oxygen Migration with Topotactic Phase Transition in Perovskite Oxides

Transition-metal perovskite oxides constitute a series of functional material systems for electronics, catalysis and energy-conversion processes, in which oxygen migration and evolution play a key role. However, the stable metal-oxygen (M-O) bond forms large energy barrier inhibiting ion diffusion. Therefore, seeking efficient and facile approaches to accelerate oxygen kinetics has become a significant issue. Here, the interaction (interfacial charge transfer and cooperative bonding) between noble metal (Pt, Ag) and perovskites oxide (SrCoO3-delta) is employed to weaken (M-O) bond and decrease the energy barrier of oxygen migration. Noble metal layers serving as oxygen pumps can continuously extract oxygen from oxide films to atmosphere. The temperature of topotactic phase transition from perovskite (SrCoO3) to brownmillerite (SrCoO2.5) is remarkably lowered from 200 K to room temperature. Furthermore, this approach can also be applied to SrFeO3 for similar topotactic phase reduction by moderate thermal activation. Our finding paves a promising and general pathway to achieve fast oxygen migration in perovskite oxides, with important application prospect in low-temperature electrodes and high-activity catalysis interface.

cond-mat.mtrl-sci↗

arXiv2021

Interface-enhanced ferromagnetism with long-distance effect in van der Waals semiconductor

Ferromagnetic semiconductors discovered in two-dimensional (2D) materials open an avenue for highly integrated and multifunctional spintronics. The Curie temperature (TC) of existed 2D ferromagnetic semiconductors is extremely low and the modulation effect of their magnetism is limited compared with their 2D metallic counterparts. The interfacial effect was found to effectively manipulate the three-dimensional magnetism, providing a unique opportunity for tailoring the 2D magnetism. Here we demonstrate that the TC of 2D ferromagnetic semiconductor Cr2Ge2Te6 can be enhanced by 130% (from ~65 K to above 150 K) when adjacent to a tungsten layer. The interfacial W-Te bonding contributes to the TC enhancement with a strong perpendicular magnetic anisotropy (PMA), guaranteeing an efficient magnetization switching by the spin-orbit torque with a low current density at 150 K. Distinct from the rapid attenuation in conventional magnets, the interfacial effect exhibits a weak dependence on Cr2Ge2Te6 thickness and a long-distance effect (more than 10 nanometers) due to the weak interlayer coupling inherent to 2D magnets. Our work not only reveals a unique interfacial behavior in 2D materials, but also advances the process towards practical 2D spintronics.

cond-mat.mtrl-sci↗

arXiv2018

Learning Shaping Strategies in Human-in-the-loop Interactive Reinforcement Learning

Providing reinforcement learning agents with informationally rich human knowledge can dramatically improve various aspects of learning. Prior work has developed different kinds of shaping methods that enable agents to learn efficiently in complex environments. All these methods, however, tailor human guidance to agents in specialized shaping procedures, thus embodying various characteristics and advantages in different domains. In this paper, we investigate the interplay between different shaping methods for more robust learning performance. We propose an adaptive shaping algorithm which is capable of learning the most suitable shaping method in an on-line manner. Results in two classic domains verify its effectiveness from both simulated and real human studies, shedding some light on the role and impact of human factors in human-robot collaborative learning.

cs.HC↗