Search

Real-time speech interaction, serving as a fundamental interface for human-machine collaboration, holds immense potential. However, current open-source models face limitations such as high costs in voice data collection, weakness in dynamic control, and limited intelligence. To address these challenges, this paper introduces Step-Audio, the first production-ready open-source solution. Key contributions include: 1) a 130B-parameter unified speech-text multi-modal model that achieves unified understanding and generation, with the Step-Audio-Chat version open-sourced; 2) a generative speech data engine that establishes an affordable voice cloning framework and produces the open-sourced lightweight Step-Audio-TTS-3B model through distillation; 3) an instruction-driven fine control system enabling dynamic adjustments across dialects, emotions, singing, and RAP; 4) an enhanced cognitive architecture augmented with tool calling and role-playing abilities to manage complex tasks effectively. Based on our new StepEval-Audio-360 evaluation benchmark, Step-Audio achieves state-of-the-art performance in human evaluations, especially in terms of instruction following. On open-source benchmarks like LLaMA Question, shows 9.3% average performance improvement, demonstrating our commitment to advancing the development of open-source multi-modal language technologies. Our code and models are available at https://github.com/stepfun-ai/Step-Audio.

We theoretically investigate the magnetic excitations in the quantum anomalous Hall insulator phase of twisted bilayer MoTe$_2$ at a hole filling factor of $\nu=1$, focusing on magnon and domain wall excitations. Using a generalized interacting Kane-Mele model, we obtain the quantum anomalous Hall insualtor ground state with spin polarization. The magnon spectrum is then computed via the Bethe-Salpeter equation, revealing two low-energy topological magnon bands with opposite Chern numbers. To further explore the magnon topology, we construct a tight-binding model for the magnon bands, which is analogous to the Haldane model. We also calculate the energy cost of domain walls that separate regions with opposite Chern numbers and bind chiral edge states. Finally, we propose an effective spin model that describes both magnon and domain wall excitations, incorporating Heisenberg spin interactions and Dzyaloshinskii-Moriya interactions. The coupling constants in this model are determined from the numerical results for magnons and domain walls. This model accounts for the Ising anisotropy of the system, captures the magnon topology, and allows for the estimation of the magnetic ordering temperature. Our findings provide a comprehensive analysis of magnetic excitations in twisted MoTe$_2$ and offer new insights into collective excitations in moir\'e systems., Comment: 13 pages,5 figures

The search for the compact limit of multi-quark states is a challenging issue. Within the framework of the MIT bag model, we propose an effective limit bag radius of $R_{c} = 5.615 \, \text{GeV}^{-1}$(or $1.11 \, \text{fm}$) for bound states . When the bag radius of a hadron falls below this value, the bag binding energy satisfies $E_{B} < 0$, indicating that the system has a compact intention. We consider various combinations of different numbers and ratios of heavy and light quarks, indicating that the bag radius of hadrons depending on the number of quarks is suppressed by the presence of heavy quarks. Focusing on five-quark combinations, we find that the average bag radius of $nncc\bar{c}$ is below the threshold $R_{c}$. We take into account color-magnetic interactions and calculate the mass, magnetic moment, binding energy, and relative strong decay width for the $nnQQ\bar{Q}$ system. We show that the binding energies of most states in the flavor combination $nncc\bar{c}$ are approximately $-20 \, \text{MeV}$, whereas states involving bottom quarks have binding energies around $-120 \, \text{MeV}$, with some decay widths suppressed by the decay constant. Additionally, the $nQQQ\bar{Q}$ system exhibits even deeper binding. Our results support the compact intention of $nnQQ\bar{Q}$ and suggest that the $nQQQ\bar{Q}$ configuration demonstrates even stronger compactness., Comment: 15 pages,2 figs

Intervalley coherent (IVC) phase in graphene systems arises from the coherent superposition of wave functions of opposite valleys, whose direct microscopic visualization provides pivotal insight into the emergent physics but remains elusive. Here, we successfully visualize the IVC phase in a heterostructure of monolayer PtSe2 on highly oriented pyrolytic graphite. Using spectroscopic imaging scanning tunneling microscopy, we observe a Root3 by Root3 modulation pattern superimposed on the higher-order moire superlattice of the heterostructure, which correlates with a small gap opening around the Fermi level and displays an anti-phase real-space conductance distribution of the two gap edges. Such modulation pattern and small-gap vanish on the heterostructure of monolayer PtSe2 on bilayer-graphene-covered SiC substrate, due to the increased carrier density in the bilayer graphene. We provide a theoretical mechanism that the Root3 by Root3 modulation pattern originates from the IVC phase of few-layer graphene, which is magnified by the higher-order moire superlattice. Our work achieves visualization of the IVC phase, and develops an avenue for its generation and amplification via a moir\'e interface., Comment: 16 pages, 4 figures

In brane-world theory, through Kaluza-Klein (KK) reduction, a higher-dimensional U(1) gauge field manifests on the brane as a series of vector and scalar KK modes, while a bulk fermion field manifests as left- and right-handed components. However, these conclusions rely on the common assumption that there is no interaction between different levels of KK modes. Recent experimental phenomena, such as flavor mixing in particles, suggest that such interactions should be taken into account. To address this, we propose an \emph{Orthonormal Completeness Hypothesis} (OCH) for the basis functions used to expand the higher-dimensional field. By applying the OCH, we demonstrate that the effective action of a free bulk U(1) gauge field is intrinsically gauge-invariant in brane models with codimension-\(d\) (\(d \geq 1\)). This effective action suggests the existence of interactions between different levels of KK modes, which can only be eliminated by choosing specific basis functions, provided the warp factors satisfy special commutation relations. In general, such interactions are universally present. We show the numerical calculations for these coupling coefficients in an interesting 6D brane world. This method can be extended to fermion fields, and it is shown that interactions between different levels of left- and right-handed KK modes exist, providing new insights into phenomena such as flavor mixing., Comment: 31 pages, 6 figures, 2 tables

Unsupervised Domain Adaptation (UDA) aims to adapt a model trained on a labeled source domain to an unlabeled target domain by addressing the domain shift. Existing Unsupervised Domain Adaptation (UDA) methods often fall short in fully leveraging contextual information from the target domain, leading to suboptimal decision boundary separation during source and target domain alignment. To address this, we introduce GrabDAE, an innovative UDA framework designed to tackle domain shift in visual classification tasks. GrabDAE incorporates two key innovations: the Grab-Mask module, which blurs background information in target domain images, enabling the model to focus on essential, domain-relevant features through contrastive learning; and the Denoising Auto-Encoder (DAE), which enhances feature alignment by reconstructing features and filtering noise, ensuring a more robust adaptation to the target domain. These components empower GrabDAE to effectively handle unlabeled target domain data, significantly improving both classification accuracy and robustness. Extensive experiments on benchmark datasets, including VisDA-2017, Office-Home, and Office31, demonstrate that GrabDAE consistently surpasses state-of-the-art UDA methods, setting new performance benchmarks. By tackling UDA's critical challenges with its novel feature masking and denoising approach, GrabDAE offers both significant theoretical and practical advancements in domain adaptation.

Motivated by the recent discovery of superconductivity in twisted bilayer WSe$_2$, we theoretically investigate the pairing instability induced by Coulomb interaction via the Kohn-Luttinger mechanism. Within random phase approximation, we show that the Coulomb repulsion is strongly screened around the mori\'e length scale, resulting in attractive regimes that facilitate electron pairing in unconventional channels. By solving the linearized gap equation, we find that the maxima of the superconducting critical temperature occur at the van Hove singularities in the density of states, with magnitude on the order of $\sim$0.2 K, comparable with experimental observations. Moreover, we identify a phase transition from an $f$-wave-like pairing state to a doubly degenerate $d$-wave-like pairing state upon increasing the electric displacement field. These findings provide a purely electronic explanation of the superconducting state observed in twisted bilayer WSe$_2$, offering insights into the interplay between superconductivity and other correlated states in twisted transition metal dichalcogenides., Comment: 7+7 pages, 5+6 figures

Flat bands and electron correlation in moir\'e lattices give rise to many exotic phases, including Mott insulators, superconductivity, and topological states. Within the first moir\'e band, integer and fractional quantum anomalous Hall effects have been observed in twisted bilayer MoTe2 (tMoTe2) at one hole doping and fractional doping per moir\'e unit cell, respectively. When the second moir\'e band is fully hole doped, quantum spin Hall insulator has also been reported in tMoTe2 at a certain twist angle. Exotic topological states together with ferromagnetic (FM) states in the high moir\'e band can potentially exist as well. In this study, we report the observation of a FM phase in the second moir\'e band in tMoTe2. The FM phase can be tuned by both the doping level and displacement field. At filling around 2.58 holes per moir\'e unit cell, the FM phase reaches a Curie temperature of 3.5 K. A large displacement field can suppress the FM phase, like the FM phase at the filling of -1. Our results demonstrate the realization of time-reversal symmetry-breaking states in the higher moir\'e bands in tMoTe2., Comment: Main text: 13 pages, 5 figures. Supplementary: 11 pages, 15 figures

We theoretically derive the sum rule for the negative first moment of the absorptive optical conductivity with excitonic effects and establish its connection to the quantum weight $K$ and Chern number $C$ of the ground state. Applying this framework, we investigate the excitonic optical response of the Chern insulator at hole filling factor $\nu=1$ in twisted bilayer MoTe$_2$. A single chiral exciton state, which selectively absorbs circularly polarized light of a specific handedness, dominates the optical sum rule. The chiral exciton state comprises two types of interlayer electron-hole transitions, which cancel out the total out-of-plane dipole moment. The absorption spectrum shows nearly perfect magnetic circular dichroism, which can be attributed to the nearly saturated bound $K \ge |C|$ of the Chern insulator under study. Our work illustrates the potential of using excitonic optical responses to probe quantum geometry encoded by $K$ and $C$ of Chern insulators in moir\'e superlattices., Comment: 7+10 pages,4+7 figures

Motivated by recent experimental observations of displacement-field-tuned correlated insulators at integer and half-integer fillings in chirally twisted triple bilayer graphene (CTTBG), we study the single-particle and interacting physics of CTTBG. We find that there are two inequivalent stacking orders, {\it i.e.}, ABABBC and ABABAB, and both exhibit flat bands with nontrivial topology. We then use the Hartree-Fock approximation to calculate the rich phase diagram of CTTBG at all integer and half-integer fillings in both stacking orders and under the vertical displacement field. Under a small displacement field, the groundstates are flavor polarized states for ABABBC stacking order and intervalley coherent states for ABABAB stacking order at all integer and half-integer fillings. A larger displacement field will turn them into layer-polarized states. At half-integer fillings, the groundstates also exhibit charge density wave (CDW) order. For ABABAB stacking, the groundstates are always $2\times1$ stripe state among a range of displacement fields. For ABABBC stacking, the groundstates are also $2\times1$ stripe states under a small displacement field and a larger displacement will possibly favor further translation-symmetry-breaking, depending on filling and the direction of the displacement field. We demonstrate that the CDW states observed in the experiment can originate from the strong Coulomb interaction of the flat band electrons.

Moir\'e superlattices have become an emergent solid-state platform for simulating quantum lattice models. However, in single moir\'e device, Hamiltonians parameters like lattice constant, hopping and interaction terms can hardly be manipulated, limiting the controllability and accessibility of moire quantum simulator. Here, by combining angle-resolved photoemission spectroscopy and theoretical analysis, we demonstrate that high-order moir\'e patterns in graphene-monolayered xenon/krypton heterostructures can simulate honeycomb model in mesoscale, with in-situ tunable Hamiltonians parameters. The length scale of simulated lattice constant can be tuned by annealing processes, which in-situ adjusts intervalley interaction and hopping parameters in the simulated honeycomb lattice. The sign of the lattice constant can be switched by choosing xenon or krypton monolayer deposited on graphene, which controls sublattice degree of freedom and valley arrangment of Dirac fermions. Our work establishes a novel path for experimentally simulating the honeycomb model with tunable parameters by high-order moir\'e patterns., Comment: 19 pages, 5 figure

In this work, we study masses and decays of triply-heavy pentaquarks $QQQn\bar{n}(Q=b,c)$ in the unified MIT bag model. We construct the color-spin wave functions of the triply-heavy pentaquarks we address and use numerical variational method to compute all ground-state masses of these system. By excluding the scattering states in these configurations, we compute the decay width ratios of each decay channels relative to the maximum width for the compact pentaquark states, obtaining main decay modes of the triply-heavy pentaquark systems., Comment: 12 pages, 10 tables, 3 figures and two appendices, to be published in Chinese Physics C

In light of the recently observed resonance states $P_{\psi s}^{\Lambda}(4338)^0$ and $P_{cs}(4459)^0$ by LHCb Collaboration in $J/\psi\Lambda$ decay channel, we perform a systematical study of all possible hidden-heavy pentaquarks with strangeness $S=0,-1,-2,-3$, in unified framework of MIT bag model. The color-spin wavefunctions presented in terms of Young-Yamanouchi bases and transformed into baryon-meson couplings, are utilized to calculate masses, magnetic moments and ratios of partial widths. With numerical analysis, the observed $P_{\psi s}^{\Lambda}(4338)^0$ is likely to be a $1/2^-$ compact $P_{cs}$ pentaquark, and $P_{cs}(4459)^0$ favors two-peak structure of $3/2^-$ and $1/2^-$ $P_{cs}$ states. Further predictions on hadron properties and decay channels are given to compact $P_{css}$, $P_{csss}$ states and bottom sectors., Comment: 14 pages, 11 tables, 1 figure and two appendices, created by RevTex 4

Consider a communication system in which a single antenna user equipment exchanges information with a multi-antenna base station via a reconfigurable intelligent surface (RIS) in the presence of spatially correlated channels and electromagnetic interference (EMI). To exploit the attractive advantages of RIS technology, accurate configuration of its reflecting elements is crucial. In this paper, we use statistical knowledge of channels and EMI to optimize the RIS elements for i) accurate channel estimation and ii) reliable data transmission. In both cases, our goal is to determine the RIS coefficients that minimize the mean square error, resulting in the formulation of two non-convex problems that share the same structure. To solve these two problems, we present an alternating optimization approach that reliably converges to a locally optimal solution. The incorporation of the diagonally scaled steepest descent algorithm, derived from Newton's method, ensures fast convergence with manageable complexity. Numerical results demonstrate the effectiveness of the proposed method under various propagation conditions. Notably, it shows significant advantages over existing alternatives that depend on a sub-optimal configuration of the RIS and are derived on the basis of different criteria., Comment: 13 pages, 10 figures

In this work, we study masses and other static properties of triply heavy tetraquarks in the unified framework of the MIT bag which incorporates chromomagnetic interactions and enhanced binding energy. The masses, magnetic moments and charge radii of all strange and nonstrange (ground) states of triply heavy tetraquarks are computed, suggesting that all of triply heavy tetraquarks are above the respective two-meson thresholds. We also estimate relative decay widths of main decay channels of two-heavy mesons for these tetraquarks., Comment: 19 pages, 6 figures, 5 tables and two appendices, created by RevTex 4

We present a theoretical study of an interaction-driven quantum phase diagram of twisted bilayer MoTe$_2$ at hole filling factor $\nu_h=1$ as a function of twist angle $\theta$ and layer potential difference $V_z$, where $V_z$ is generated by an applied out-of-plane electric field. At $V_z=0$, the phase diagram includes quantum anomalous Hall insulators in the intermediate $\theta$ regime and topologically trivial multiferroic states with coexisting ferroelectricity and magnetism in both small and large $\theta$ regimes. There can be two transitions from the quantum anomalous Hall insulator phase to topologically trivial out-of-plane ferromagnetic phase, and finally to in-plane 120$^\circ$ antiferromagnetic phase as $|V_z|$ increases, or a single transition without the intervening ferromagnetic phase. We show explicitly that the spin vector chirality of various 120$^\circ$ antiferromagnetic states can be electrically switched. We discuss the connection between the experimentally measured Curie-Weiss temperature and the low-temperature magnetic order based on an effective Heisenberg model with magnetic anisotropy., Comment: 7 pages, 4 figures

Semiconductor moir\'e superlattices provide a highly tunable platform to study the interplay between electron correlation and band topology. For example, the generalized Kane-Mele-Hubbard model can be simulated by the topological moir\'e flat bands in twisted transition metal dichalcogenides homobilayers. For this system, we obtain the filling factor, twist angle, and electric field-dependent quantum phase diagrams with a plethora of phases, including the quantum spin Hall insulator, the in-plane antiferromagnetic state, the out-of-plane antiferromagnetic Chern insulator, the spin-polarized Chern insulator, the in-plane ferromagnetic state, and the 120$^\circ$ antiferromagnetic state. We predict that a gate-defined junction formed between the quantum spin Hall insulator phase with proximitized superconductivity and magnetic phases with in-plane magnetization (either ferromagnetic or antiferromagnetic) can realize one-dimensional topological superconductor with Majorana zero modes. Our proposal introduces semiconductor moir\'e homobilayers as an electrically tunable Majorana platform with no need of an external magnetic field., Comment: 12 pages, 9 figures

Twisted bilayer MoTe$_2$ is a promising platform to investigate the interplay between band topology and many-body interaction. We present a theoretical study of its interaction-driven quantum phase diagrams based on a three-orbital model, which can be viewed as a generalization of the Kane-Mele-Hubbard model with one additional orbital and long-range Coulomb repulsion. We predict a cascade of phase transitions tuned by the twist angle $\theta$. At the hole filling factor $\nu=1$ (one hole per moir\'e unit cell), the ground state can be in the multiferroic phase with coexisting spontaneous layer polarization and magnetism, the quantum anomalous Hall phase, and finally the topologically trivial magnetic phases, as $\theta$ increases from $1.5^{\circ}$ to $5^{\circ}$. At $\nu=2$, the ground state can have a second-order phase transition between an antiferromagnetic phase and the quantum spin Hall phase as $\theta$ passes through a critical value. The dependence of the phase boundaries on model parameters such as the gate-to-sample distance, the dielectric constant, and the moir\'e potential amplitude is examined. The predicted phase diagrams can guide the search for topological phases in twisted transition metal dichalcogenide homobilayers., Comment: 16 pages, 13 figures

Inspired by the observation of the $X(6900)$ by LHCb and the $X(6600)$ (with mass $6552\pm 10$ $\pm 12$ MeV) recently by CMS and ATLAS experiments of the LHC in the di-$J/\Psi $ invariant mass spectrum, we systemically study masses of all ground-state configurations of the hidden heavy-flavor tetraquarks $q_{1}Q_{2}\bar{q}_{3}\bar{Q}_{4}$ and $Q_{1}Q_{2}\bar{Q}_{3}\bar{Q}_{4}$ ($Q=c,b$;$q=u,d,s$) contaning two and four heavy quarks in the MIT bag model with chromomagnetic interaction and enhanced binding energy. Considering color-spin mixing due to chromomagnetic interaction, our mass computation indicates that the observed $X(6600)$ is likely to be the $0^{++}$ ground states of hidden-charm tetraquark $cc\bar{c}\bar{c}$ with computed masses $6572$ MeV, which has a $0^{++}$ color partner around $6469$ MeV. The fully bottom system of tetraquark $bb\bar{b}\bar{b}$ has masses of 19685 MeV and 19717 MeV for the the $0^{++}$ ground states. Further computation is given to the tetraquark systems $sc\bar{s}\bar{c}$, $sb\bar{s}\bar{b}$, $cb\bar{c}\bar{b}$, $nc\bar{n}\bar{c}$ and $nb\bar{n}\bar{b}$, suggesting that the $Z_{c}(4200)$ is the tetraquark $nc\bar{n}\bar{c}$ with $J^{PC}=1^{+-}$. All of these tetraquarks are above their lowest thresholds of two mesons and unstable against the strong decays.

We introduce an enhanced binding energy $B_{QQ}$ between heavy-heavy quarks $QQ$ and a flux-tube correction into the chromomagnetic interaction model to study nonstrange doubly-heavy tetraquarks $T_{QQ}$ ($Q=c,b)$. A simple relation in terms of baryon masses is proposed to estimate the binding energies $B_{QQ}$ and thereby map the flux-tube corrections in doubly-heavy tetraquarks $T_{cc}$, $T_{bb}$ and $T_{bc}$. Our computation via diagonalization of chromomagnetic interaction predicts the doubly charmed tetraquark $T_{cc}$ (in color rep. $\bar{3}_{c}\otimes 3_{c}$) and $T_{cc}^{\ast }$ (in $6_{c}\otimes \bar{6}_{c}$) with $IJ^{P}=01^{+}$ to have masses of $3879.2$ MeV and $4287.6$ MeV, respectively, with the former being in consistent with the measured mass $3874.7\pm 0.05$ MeV of the doubly charmed tetraquark $T_{cc}(1^{+})=cc\bar{u}\bar{d}$ discovered by LHCb. Further mass predictions are given of the doubly bottom tetraquarks $T_{bb}$ and the bottom-charmed tetraquarks $T_{bc}$ with $J^{P}=0^{+},1^{+},2^{+}$ and $I=0,1$. A chromomagnetical mixing between the color configurations $\bar{3}_{c}\otimes 3_{c}$ and $6_{c}\otimes \bar{6}_{c}$ is noted for the bottom-charmed states $T_{bc}$ with $IJ^{P}=01^{+}$ and $IJ^{P}=1(0^{+},1^{+})$., Comment: 16 pages, 1 figures, Created by RevTex 4