SearcharxivSearch

arXiv subjects

Xiangshui Miao

Publications and source records attributed to Xiangshui Miao.

18 recordsLinked to original sources

SwiftKV: An Edge-Oriented Attention Algorithm and Multi-Head Accelerator for Fast, Efficient LLM Decoding

Edge acceleration for large language models is crucial for their widespread application; however, achieving fast attention inference and efficient decoding on resource-constrained edge accelerators remains challenging. This paper presents SwiftKV Attention, a per-token pipelined, low-latency single-pass attention inference algorithm, where every (kt, vt) in the KV cache is processed exactly once in a uniform per-token pipeline without score materialization, blockwise softmax, or a second pass, thereby enabling fast execution on edge accelerators with a single hardware set and no resource-intensive parallelism. Furthermore, to address the limited support for multi-head LLM decoding in existing accelerators, we design the SwiftKV-MHA accelerator, which enables high precision attention and low precision GEMV on the same processor array, achieving fast and efficient multi-head parallel decoding. Experimental results show that, on the edge accelerator, the SwiftKV Attention algorithm achieves a 7.16* speedup over native attention and significantly outperforms other attention algorithms. SwiftKV-MHA further reduces attention latency by 13.48*; under the same settings, it improves generation speed by 17.4% and increases token efficiency by 1.98* compared with state-of-the-art works.

cs.AR

MemIntelli: A Generic End-to-End Simulation Framework for Memristive Intelligent Computing

Memristive in-memory computing (IMC) has emerged as a promising solution for addressing the bottleneck in the Von Neumann architecture. However, the couplingbetweenthecircuitandalgorithm in IMC makes computing reliability susceptible to non-ideal effects in devices and peripheral circuits. In this respect, efficient softwarehardwareco-simulationtoolsarehighlydesiredtoembedthedevice and circuit models into the algorithms. In this paper, for the first time, we proposed an end-to-end simulation framework supporting flexible variable-precision computing, named MemIntelli, to realize the pre-verification of diverse intelligent applications on memristive devices. At the device and circuit level, mathematical functions are employed to abstract the devices and circuits through meticulous equivalent circuit modeling. On the architecture level, MemIntelli achieves flexible variable-precision IMC supporting integer and floating data representation with bit-slicing. Moreover, MemIntelli is compatible with NumPy and PyTorch for seamless integration with applications. To demonstrate its capabilities, diverse intelligent algorithms, such as equation solving, data clustering, wavelet transformation, and neural network training and inference, were employed to showcase the robust processing ability of MemIntelli. This research presents a comprehensive simulation tool that facilitates the co-design of the IMC system, spanning from device to application.

cs.AR

The occupation dependent DFT-1/2 method

There has been a high demand in rectifying the band gap under-estimation problem in density functional theory (DFT), while keeping the computational load at the same level as local density approximation. DFT-1/2 and shell DFT-1/2 are useful attempts, as they correct the spurious electron self-interaction through the application of self-energy potentials, which pull down the valence band. Nevertheless, the self-energy potential inevitably disturbs the conduction band, and these two methods fail for semiconductors whose hole and electron are entangled in the same shell-like regions. In this work, we introduce the occupation-dependent DFT-1/2 method, where conduction band states are not subject to the additional self-energy potential disturbance. This methodology works for difficult cases such as $\text{Li}_2\text{O}_2$, $\text{Cu}_2\text{O}$ and two-dimensional semiconductors. Using a shell-like region for the self-energy potential, and allowing for downscaling of the atomic self-energy potential (with an $A$ < 1 factor), the occupation-dependent shell DFT+$A$-1/2 method yields more accurate conduction band and valence band edge levels for monolayer $\text{MoS}_2$, compared with the computationally demanding hybrid functional approach.

cond-mat.mtrl-sci

Acceleration of shell DFT-1/2 in high-throughput calculations via cutoff radii prediction

Shell DFT-1/2 is a fast band gap rectification method that is versatile for semiconductor supercell and superlattice calculations, which involves two cutoff radii that have to be optimized. Although such optimization is trivial in terms of time cost for a primitive cell, in high-throughput calculations this can be a big concern because most materials are themselves in small unit cells. The numerous optimization trials increase the computational cost to orders of magnitudes higher. In this work, we construct a regression model for the prediction of the two cutoff radii based on chemical composition and primitive cell structure. Moreover, a model for metal and insulator classification is also given, with 95.2% accuracy.

cond-mat.mtrl-sci

High-efficiency computational methodologies for electronic properties and structural characterization of Ge-Sb-Te based phase change materials

Theoretical simulation to phase change materials such as Ge-Sb-Te has suffered from two methodology issues. On the one hand, there is a lack of efficient band gap correction method for density functional theory, which is suitable for these materials in both crystalline and amorphous phases, though the computational complexity should be kept at the local density approximation level. On the other hand, analysis of the coordination number in amorphous phases relies on an integration involving the radial distribution function, which adds to the complexity. In this work, we find that the shell DFT-1/2 method offers an overall band gap accuracy for phase change materials comparable to HSE06 hybrid functional, though its computational cost is around three orders of magnitude lower. Moreover, the mixed length-angle coordination number theory enables calculating the coordination numbers in the amorphous phase directly from the structure, with definite outcomes. The two methodologies could be helpful for high throughput simulation of phase change materials.

cond-mat.mtrl-sci

On the material genome of wurtzite ferroelectrics

As the dielectric film thickness shrinks to ~10 nm, some traditional wurtzite piezoelectric materials demonstrate ferroelectricity through element doping. Among them, Sc doped AlN and Mg doped ZnO are the most famous examples. While it is widely acknowledged that the dopant atoms effectively reduce the coercive field, enabling ferroelectric polarization switching, the material genome of these wurtzite (WZ) ferroelectrics is still less understood. In this work, we analyze the features of WZ ferroelectrics, ascribing them to five-coordination (5C) ferroelectrics, which may be compared with 6C ferroelectrics (perovskite-type) and 7C ferroelectrics (hafnia-like). In particular, the exact reason for their adopting the hexagonal WZ structure instead of the zinc blende structure is studied. Emphasis is paid to the degree of ionicity in promoting the hexagonal arrangement, and the phenomenon of layer distance compression is discovered and explained in WZ ferroelectrics. The role of element doping in coercive field reduction is understood within this context.

cond-mat.mtrl-sci

Structure evolution path of ferroelectric hafnium zirconium oxide nanocrystals under in-situ biasing

Fluorite-type $\mathrm{HfO_2}$-based ferroelectric (FE) oxides have rekindled interest in FE memories due to their compatibility with silicon processing and potential for high-density integration. The polarization characteristics of FE devices are governed by the dynamics of metastable domain structure evolution. Insightful design of FE devices for encoding and storage necessitates a comprehensive understanding of the internal structural evolution. Here, we demonstrate the evolution of domain structures through a transient polar orthorhombic (O)-$Pmn2_1$-like configuration via $in$-$situ$ biasing on $\mathrm{TiN/Hf_{0.5}Zr_{0.5}O_2/TiN}$ capacitors within spherical aberration-corrected transmission electron microscope, combined with theoretical calculations. Furthermore, it is directly evidenced that the non-FE O-$Pbca$ transforms into the FE O-$Pca2_1$ phase under electric field, with the polar axis of the FE-phase aligning towards the bias direction through ferroelastic transformation, thereby enhancing FE polarization. As cycling progresses further, however, the polar axis collapses, leading to FE degradation. These novel insights into the intricate structural evolution path under electrical field cycling facilitate optimization and design strategies for $\mathrm{HfO_2}$-based FE memory devices.

cond-mat.mtrl-sci

Phase transitions in typical fluorite-type ferroelectrics

While ferroelectric hafnia ($\mathrm{HfO_2}$) has become a technically important material for microelectronics, the physical origin of its ferroelectricity remains poorly understood. The tetragonal $P4_2/nmc$ phase is commonly assigned as its paraelectric mother phase but has no soft mode at the Brillouin zone center. In this work, we propose that the paraelectric-ferroelectric transition in hafnia-like $Pca2_1$ ferroelectric family can be described by a $Pcca$-$Pca2_1$ transition, where the $Pcca$ mother phase will evolve into either the $Pca2_1$ ferroelectric phase or the centrosymmetric $P2_1/c$ monoclinic phase, depending on the strain conditions. The $Pcca$ phase is directly linked to both phases in the context of continuous phase transition. Hafnia is regarded as a special case of this family, in that it has accidental atomic degeneracy because all anions are oxygen. The theory is also correlated to the seven-coordination theory that explains the ferroelectricity in hafnia from a chemical perspective. In addition, the strain conditions to promote the ferroelectric phase in hafnia are discussed.

cond-mat.mtrl-sci

Landauer-QFLPS model for mixed Schottky-Ohmic contact two-dimensional transistors

Two-dimensional material-based field effect transistors (2DM-FETs) are playing a revolutionary role in electronic devices. However, after years of development, no device model can match the Pao-Sah model for standard silicon-based transistors in terms of physical accuracy and computational efficiency to support large-scale integrated circuit design. One remaining critical obstacle is the contacts of 2DM-FETs. In order to self-consistently include the contact effect in the current model, it is necessary to perform self-consistent calculations, which is a fatal flaw for applications that prioritize efficiency. Here, we report that the Landauer-QFLPS model effectively overcomes the above contradiction, where QFLPS means quasi-Fermi-level phase space theory. By connecting the physical pictures of the contact and the intrinsic channel part, we have successfully derived a drain-source current formula including the contact effect. To verify the model, we prepared transistors based on two typical 2DMs, black phosphorus (BP) and molybdenum disulfide (MoS2), the former having ambipolar transport and the latter showing electron-dominant unipolar transport. The proposed new formula could describe both 2DM-FETs with Schottky or Ohmic contacts. Moreover, compared with traditional methods, the proposed model has the advantages of accuracy and efficiency, especially in describing non-monotonic drain conductance characteristics, because the contact effect is self-consistently and compactly packaged as an exponential term. More importantly, we also examined the model at the circuit level. Here, we fabricated a three-bit threshold inverter quantizer circuit based on ambipolar-BP process and experimentally demonstrated that the model can accurately predict the circuit performance. This industry-benign 2DM-FET model is supposed to be very useful for the development of 2DM-FET-based integrated circuits.

physics.app-ph

Impact of Zr substitution on the electronic structure of ferroelectric hafnia

$\mathrm{HfO_2}$-based dielectrics are promising for nanoscale ferroelectric applications, and the most favorable material within the family is Zr-substituted hafnia, i.e., $\mathrm{Hf_{1-x}Zr_xO_2}$ (HZO). The extent of Zr substitution can be great, and x is commonly set to 0.5. However, the band gap of $\mathrm{ZrO_2}$ is lower than $\mathrm{HfO_2}$, thus it is uncertain how the Zr content should influence the electronic band structure of HZO. A reduced band gap is detrimental to the cycling endurance as charge injection and dielectric breakdown would become easier. Another issue is regarding the comparison on the band gaps between $\mathrm{HfO_2}$/$\mathrm{ZrO_2}$ superlattices and HZO solid-state solutions. In this work we systematically investigated the electronic structures of $\mathrm{HfO_2}$, $\mathrm{ZrO_2}$ and HZO using self-energy corrected density functional theory. In particular, the conduction band minimum of $Pca2_1$-$\mathrm{HfO_2}$ is found to lie at an ordinary k-point on the Brillouin zone border, not related to any interlines between high-symmetry k-points. Moreover, the rule of HZO band gap variation with respect to x has been extracted. The physical mechanisms for the exponential reduction regime and linear decay regime have been revealed. The band gaps of $\mathrm{HfO_2}$/$\mathrm{ZrO_2}$ ferroelectric superlattices are investigated in a systematic manner, and the reason why the superlattice could possess a band gap lower than that of $\mathrm{ZrO_2}$ is revealed through comprehensive analysis.

cond-mat.mtrl-sci

On the self-consistency of DFT-1/2

DFT-1/2 is an efficient band gap rectification method for density functional theory (DFT) under local density approximation (LDA) or generalized gradient approximation. It was suggested that non-self-consistent DFT-1/2 should be used for highly ionic insulators like LiF, while self-consistent DFT-1/2 should still be used for other compounds. Nevertheless, there is no quantitative criterion prescribed for which implementation should work for an arbitrary insulator, which leads to severe ambiguity in this method. In this work we analyze the impact of self-consistency in DFT-1/2 and shell DFT-1/2 calculations in insulators or semiconductors with ionic bonds, covalent bonds and intermediate cases, and show that self-consistency is required even for highly ionic insulators for globally better electronic structure details. The self-energy correction renders electrons more localized around the anions in self-consistent LDA-1/2. The well-known delocalization error of LDA is rectified, but with strong overcorrection due to the presence of additional self-energy potential. However, in non-self-consistent LDA-1/2 calculations, the electron wavefunctions indicate that such localization is much more severe and beyond a reasonable range, because the strong Coulomb repulsion is not counted in the Hamiltonian. Another common drawback of non-self-consistent LDA-1/2 lies in that the ionicity of the bonding gets substantially enhanced, and the band gap can be enormously high in mixed ionic-covalent compounds like $\mathrm{TiO_2}$. The impact of LDA-1/2-induced stress is also discussed comprehensively.

cond-mat.mtrl-sci

Designing wake-up free ferroelectric capacitors based on the $\mathrm{HfO_2/ZrO_2}$ superlattice structure

The wake-up phenomenon widely exists in hafnia-based ferroelectric capacitors, which causes device parameter variation over time. Crystallization at higher temperatures have been reported to be effective in eliminating wake-up, but high temperature may yield the monoclinic phase or generate high concentration oxygen vacancies. In this work, a unidirectional annealing method is proposed for the crystallization of $\mathrm{Hf_{0.5}Zr_{0.5}O_2}$ (HZO) superlattice ferroelectrics, which involves heating from the $\mathrm{Pt/ZrO_2}$ interface side. Nanoscale $\mathrm{ZrO_2}$ is selected to resist the formation of monoclinic phase, and the chemically inert Pt electrode can avoid the continuous generation of oxygen vacancies during annealing. It is demonstrated that $\mathrm{600^oC}$ annealing only leads to a moderate content of monoclinic phase in HZO, and the TiN/HZO/Pt capacitor exhibits wake-up free nature and a $2P_\mathrm{r}$ value of 27.4 $μ\mathrm{C/cm^2}$. On the other hand, heating from the $\mathrm{TiN/HfO_2}$ side, or using $\mathrm{500^oC}$ annealing temperature, both yield ferroelectric devices that require a wake-up process. The special configuration of $\mathrm{Pt/ZrO_2}$ is verified by comparative studies with several other superlattice structures and HZO solid-state solutions. It is discovered that heating from the $\mathrm{Pt/HfO_2}$ side at $\mathrm{600^oC}$ leads to high leakage current and a memristor behavior. The mechanisms of ferroelectric phase stabilization and memristor formation have been discussed. The unidirectional heating method can also be useful for other hafnia-based ferroelectric devices.

cond-mat.mtrl-sci

Shell DFT-1/2 method towards engineering accuracy for semiconductors: GGA versus LDA

The Kohn-Sham gaps of density functional theory (DFT) obtained in terms of local density approximation (LDA) or generalized gradient approximation (GGA) cannot be directly linked to the fundamental gaps of semiconductors, but in engineering there is a strong demand to match them through certain rectification methods. Shell DFT-1/2 (shDFT-1/2), as a variant of DFT-1/2, is a potential candidate to yield much improved band gaps for covalent semiconductors, but its accuracy depends on the LDA/GGA ground state, including optimized lattice parameters, basic Kohn-Sham gap before self-energy correction and the amount of self-energy correction that is specific to the exchange-correlation (XC) functional. In this work, we test the LDA/GGA as well as shDFT-1/2 results of six technically important covalent semiconductors Si, Ge, GaN, GaP, GaAs and GaSb, with an additional ionic insulator LiF for comparison. The impact of XC flavor (LDA, PBEsol, PBE and RPBE), either directly on the gap value, or indirectly through the optimized lattice constant, is examined comprehensively. Moreover, we test the impact of XC flavor on LDA/GGA and shDFT-1/2 gaps under the condition of fixed experimental lattice constants. In-depth analysis reveals the rule of reaching the best accuracy in calculating the electronic band structures of typical covalent semiconductors. Relevant parameters like lattice constant, self-consistency in shDFT-1/2 runs, as well as the exchange enhancement factor of GGA, are discussed in details.

cond-mat.mtrl-sci

Hafnia for analog memristor: Influence of stoichiometry and crystalline structure

The highly non-linear switching behavior of hafnia memristor actually hinders its wide application in neuromorphic computing. Theoretical understanding into its switching mechanism has been focused on the processes of conductive filament generation and rupture, but possible phase transition and crystallization around the region of conductive filaments (CFs) due to the variation of O content have been paid less attention to. In this paper, HfO$\mathrm{_x}$ structural models covering the full stoichiometries from Hf to HfO$\mathrm{_2}$ were established, and the crystal structure evolution during the reduction process of hafnia was obtained through first-principles calculation. The electronic structures and O vacancy migration characteristics of these structures were analyzed. A criterion was prescribed to predict the mode of abrupt binary switching or gradual conductance modulation according to the structure evolution of the CFs. In particular, factors that influence the merging of tiny conductive channels into strong filaments are intensively discussed, including the anisotropy of O vacancy migration and the size effect. The feasibility of Mg doping to achieve robust gradual switching is discussed.

cond-mat.mtrl-sci

Ferroelectricity in $\mathrm{HfO_2}$ from a chemical perspective

Ferroelectricity observed in thin film $\mathrm{HfO_2}$, either doped with Si, Al, etc. or in the $\mathrm{Hf_{0.5}Zr_{0.5}O_2}$ form, has gained great technical significance. However, the soft mode theory faces a difficulty in explaining the origin of such ferroelectricity. In this work, we propose that the 7 cation coordination number of $\mathrm{HfO_2/ZrO_2}$ lies at the heart of this ferroelectricity, which stems from the proper ionic radii of Hf/Zr compared with O. Among the numerous compounds with non-centrosymmetric nature, e.g., $mm2$ point group, $\mathrm{HfO_2}$ and $\mathrm{ZrO_2}$ are special in that they are close to the border of 7 and 8 cation coordination, such that the 8-coordination tetragonal intermediate phase could greatly reduce the switching barrier. Other 7-coordination candidates, including $\mathrm{SrI_2}$, TaON, YSBr and YOF are also studied in comparison to $\mathrm{HfO_2}$/$\mathrm{ZrO_2}$, and six switching paths are analyzed in detail for the $Pca2_1$ phase. A rule of preferred switching path in terms of ionic radii ratio and coordination number has been established. We also show the possible route from ferroelectric $Pca2_1$ phase to monoclinic $P2_1/c$ phase in $\mathrm{HfO_2}$, which is relevant to the fatigue phenomenon.

cond-mat.mtrl-sci

Isolating hydrogen in hexagonal boron nitride bubbles by a plasma treatment

Atomically thin hexagonal boron nitride (h-BN) is often regarded as an elastic film that is impermeable to gases. The high stabilities in thermal and chemical properties allow h-BN to serve as a gas barrier under extreme conditions.In this work, we demonstrate the isolation of hydrogen in bubbles of h-BN via plasma treatment.Detailed characterizations reveal that the substrates do not show chemical change after treatment. The bubbles are found to withstand thermal treatment in air,even at 800 degree celsius. Scanning transmission electron microscopy investigation shows that the h-BN multilayer has a unique aligned porous stacking nature, which is essential for the character of being transparent to atomic hydrogen but impermeable to hydrogen molecules. We successfully demonstrated the extraction of hydrogen gases from gaseous compounds or mixtures containing hydrogen element. The successful production of hydrogen bubbles on h-BN flakes has potential for further application in nano/micro-electromechanical systems and hydrogen storage.

physics.app-ph

Pressure-Induced Structural Phase Transition and a Special Amorphization Phase of Two-Dimensional Ferromagnetic Semiconductor Cr2Ge2Te6

Layered transition-metal trichalcogenides have become one of the research frontiers as two-dimensional magnets and candidate materials used for phase-change memory devices. Herein we report the high-pressure synchrotron X-ray diffraction and resistivity measurements on Cr2Ge2Te6 (CGT) single crystal by using diamond anvil cell techniques, which reveal a mixture of crystalline-to-crystalline and crystalline-to-amorphous transitions taking place concurrently at 18.3-29.2 GPa. The polymorphic transition could be interpreted by atomic layer reconstruction and the amorphization could be understood in connection with randomly flipping atoms into van der Waals gaps. The amorphous (AM) phase is quenchable to ambient conditions. The electrical resistance of CGT shows a bouncing point at ~ 18 GPa, consistent with the polymorphism phase transition. Interestingly, the high-pressure AM phase exhibits metallic resistance with the magnitude comparable to that of high-pressure crystalline phases, whereas the resistance of the AM phase at ambient pressure fails to exceed that of the crystalline phase, indicating that the AM phase of CGT appeared under high pressure is quite unique and similar behavior has never been observed in other phase-change materials. The results definitely would have significant implications for the design of new functional materials.

cond-mat.str-el

Nitrogen-induced Local Spin Polarization in Graphene on Cobalt

By means of first principles calculations we demonstrate an effective method to tailor the local spin configuration of graphene on Co(0001) surface through nitrogen doping. Two different site occupancies of the N impurities are discussed with the focus on structural, electronic and magnetic properties. N induces opposite local spin polarization at the two sites through Zener exchange-type hybridization with Co substrate. In addition, the induced spin polarization is energy dependent and controllable by electric field. Consequently this structure can be applied as a spin injection source in graphene based spintronics.

cond-mat.mtrl-sci