Search

An emerging commonality among the recently observed pulsar halos is the presence of distinct radiation patterns at high energies, while no extended radiation is detected around the GeV energy band. This commonality suggests that pulsar halos play a crucial role in the local propagation of cosmic rays, making it necessary to investigate the underlying mechanisms of this phenomenon. This work focuses on the 3D propagation study of cosmic rays, incorporating the Geminga pulsar into our propagation model to investigate its contribution to different observational spectra. We consider Geminga a dominant local source of positrons, successfully reproducing the observed positron spectrum and multi-energy spectra of the Geminga halo. Through calculations of signal and background at different angles, we find that: (1) at low energies, the slow diffusion characteristic around the pulsar region leads to a low electron density in the extended area around Geminga, causing the background radiation to exceed the signal intensity far; (2) at high energies, the larger effective diffusion radius of high-energy electrons/positrons causes the signal from Geminga to dominate the local high-energy phenomena; (3) the observed fluctuation of diffuse gamma-ray radiation by LHAASO is likely due to the incomplete subtraction of radiation from the local halo. We hope LHAASO will detect more cosmic ray halo sources to validate our model further.

The origin of the binary black hole mergers observed by LIGO-Virgo-KAGRA (LVK) remains an open question. We calculate the merger rate from primordial black holes (PBHs) within the density spike around supermassive black holes (SMBHs) at the center of galaxies. We show that the merger rate within the spike is comparable to that within the wider dark matter halo. We also calculate the extreme mass ratio inspiral (EMRI) signal from PBHs hosted within the density spike spiralling into their host SMBHs due to GW emission. We predict that LISA may detect $\sim10^4$ of these EMRIs with signal-to-noise ratio of 5 within a 4-year observation run, if all dark matter is made up of PBHs. Uncertainties in our rates come from the uncertain mass fraction of PBHs within the dark matter spike, relative to the host central SMBHs, which defines the parameter space LISA can constrain., Comment: 6 pages, 3 figures, plus Appendix (3 figures, 1 table)

Given the vast differences in interaction details, describing the dynamics of structurally disordered materials in a unified theoretical framework presents a fundamental challenge to condensed-matter physics and materials science. This paper investigates numerically a percolation scenario for the two most important relaxation processes of supercooled liquids and glasses. For nine binary glass formers we find that, as temperature is lowered from the liquid state, percolation of immobile particles takes place at the temperature locating the $\alpha$ process. Mirroring this, upon continued cooling into the glass, mobile-particle percolation pinpoints a Johari-Goldstein $\beta$ relaxation whenever it is well separated from the $\alpha$ process. For 2D systems under the same conditions, percolation of mobile and immobile particles occurs nearly simultaneously and no $\beta$ relaxation can be identified. Our findings suggest a general description of glassy dynamics based on a percolation perspective., Comment: Accepted by Nature Physics, "in principle" (this is the version originally submitted to NP)

Full-duplex spoken dialogue systems significantly advance over traditional turn-based dialogue systems, as they allow simultaneous bidirectional communication, closely mirroring human-human interactions. However, achieving low latency and natural interactions in full-duplex dialogue systems remains a significant challenge, especially considering human conversation dynamics such as interruptions, backchannels, and overlapping speech. In this paper, we introduce a novel End-to-End GPT-based model OmniFlatten for full-duplex conversation, capable of effectively modeling the complex behaviors inherent to natural conversations with low latency. To achieve full-duplex communication capabilities, we propose a multi-stage post-training scheme that progressively adapts a text-based large language model (LLM) backbone into a speech-text dialogue LLM, capable of generating text and speech in real time, without modifying the architecture of the backbone LLM. The training process comprises three stages: modality alignment, half-duplex dialogue learning, and full-duplex dialogue learning. Throughout all training stages, we standardize the data using a flattening operation, which allows us to unify the training methods and the model architecture across different modalities and tasks. Our approach offers a straightforward modeling technique and a promising research direction for developing efficient and natural end-to-end full-duplex spoken dialogue systems. Audio samples of dialogues generated by OmniFlatten can be found at this web site (https://omniflatten.github.io/)., Comment: Work in progress

We used FAST to conduct deep HI imaging of the entire M 106 group region, and have discovered a few new HI filaments and clouds. Three HI clouds/filaments are found in a region connecting DDO 120 and NGC 4288, indicating an interaction between these two galaxies. The HI features in this region suggest that DDO 120 is probably the origin of the HI stream extending from the northern end of NGC 4288 to M 106. This structure is similar to the SMC-LMC stream, but much longer, about 190 kpc. Furthermore, based on the distance measurements, we have determined the satellite galaxy members of M 106. With an absolute magnitude cutoff of M_B=-10, we obtained a sample of 11 member satellite galaxies for M 106. Using the observed HI mass with FAST, we studied the properties of satellite galaxies in M 106 and found that satellite galaxies with lower stellar masses exhibit more significant deviations from the star-forming main sequence (SFMS) in their specific star formation rates. Furthermore, the relationship between the HI mass of satellite galaxies and optical diameter generally follows the field galaxies relation. We discuss the possible mechanisms leading to the quenching in the M 106 group based on the new data from FAST, Comment: 18 pages,11 figures and 3 tables.Accepted by mnras

A theoretical model describing the one-dimensional large strain consolidation of the modified Cam Clay soil is presented in this paper. The model is based on the Lagrangian formulation, and is capable of featuring the variability of soil compressibility (inherently so due to the direct incorporation of the specific Cam Clay plasticity model) and permeability, as well as the impact of overconsolidation ratio. The derivation starts from the establishment of the incremental stress-strain relations for both purely elastic and elastoplastic deformations under one-dimensional compression condition, and thereafter the coefficients of compressibility/volume change that are essential to the consolidation analysis. The governing partial differential equation is then neatly deduced in conjunction with the continuity and equilibrium conditions for the soil, with the vertical effective stress being the privileged unknown to be solved for. Subsequently, semi-analytical solution to the developed rigorous poroelastoplastic large strain consolidation model is obtained and verified with the ABAQUS finite element numerical results. Parametric analyses are finally provided to investigate in detail the influences of the soil overconsolidation ratio, large strain configuration, and the variability of the soil permeability on the calculated one-dimensional consolidation response., Comment: 40 pages, 8 figures

The GD-1 stellar stream exhibits spur and gap structures that may result from a close encounter with a dense substructure. When interpreted as a dark matter subhalo, the perturber is denser than predicted in the standard cold dark matter (CDM) model. In self-interacting dark matter (SIDM), however, a halo could evolve into a phase of gravothermal collapse, resulting in a higher central density than its CDM counterpart. We conduct high-resolution controlled N-body simulations to show that a collapsed SIDM halo could account for the GD-1 perturber's high density. We model a progenitor halo with a mass of $3\times10^8~M_\odot$, motivated by a cosmological simulation of a Milky Way analog, and evolve it in the Milky Way's tidal field. For a cross section per mass of $\sigma/m\approx30-100~{\rm cm^2~g^{-1}}$ at $V_{\rm max }\sim10~{\rm km~s^{-1}}$, the enclosed mass of the SIDM halo within the inner $10~{\rm pc}$ can be increased by more than an order of magnitude compared to its CDM counterpart, leading to a good agreement with the properties of the GD-1 perturber. Our findings indicate that stellar streams provide a novel probe into the self-interacting nature of dark matter., Comment: 10 pages, 4 figures

Automatic Speech Recognition (ASR) transcripts exhibit recognition errors and various spoken language phenomena such as disfluencies, ungrammatical sentences, and incomplete sentences, hence suffering from poor readability. To improve readability, we propose a Contextualized Spoken-to-Written conversion (CoS2W) task to address ASR and grammar errors and also transfer the informal text into the formal style with content preserved, utilizing contexts and auxiliary information. This task naturally matches the in-context learning capabilities of Large Language Models (LLMs). To facilitate comprehensive comparisons of various LLMs, we construct a document-level Spoken-to-Written conversion of ASR Transcripts Benchmark (SWAB) dataset. Using SWAB, we study the impact of different granularity levels on the CoS2W performance, and propose methods to exploit contexts and auxiliary information to enhance the outputs. Experimental results reveal that LLMs have the potential to excel in the CoS2W task, particularly in grammaticality and formality, our methods achieve effective understanding of contexts and auxiliary information by LLMs. We further investigate the effectiveness of using LLMs as evaluators and find that LLM evaluators show strong correlations with human evaluations on rankings of faithfulness and formality, which validates the reliability of LLM evaluators for the CoS2W task., Comment: 7 pages, 3 figures

We investigate the impact of galactic disks on the tidal stripping of cold dark matter subhalos within Milky Way (MW)-mass halos ($M_{\rm vir}\sim 10^{12}\mathrm{M_{\odot}}$) using a new simulation suite, EDEN. By re-simulating 45 MW-mass zoom-in halos from the N-body Symphony compilation with embedded disk potentials, which evolve according to star formation histories predicted by the UniverseMachine model, we self-consistently tie disk growth to halo accretion rate and significantly expand the range of disk masses and formation histories studied. We use the particle-tracking-based subhalo finder Symfind to enhance the robustness of subhalo tracking. We find that disks near the median disk-to-halo mass ratio of our sample ($M_{\ast, \rm Disk}/M_{\rm vir, host} = 2\%$) reduce subhalo peak mass functions within 100 kpc by about $10\%$ for peak masses above $ 10^8\mathrm{M_{\odot}}$. Heavier, MW/M31-like disks ($M_{\ast, \rm Disk}/M_{\rm vir, host} \gtrsim 5\%$) lead to a reduction of more than $40\%$. Subhalo abundance suppression is more pronounced near halo centers, with particularly enhanced stripping for subhalos accreted over 8 Gyr ago on orbits with pericenters < 100 kpc. Suppression is further amplified when disk mass is increased within fixed halo and disk assembly histories. In all cases, the suppression we measure should be interpreted as stripping below the mass resolution limit rather than complete subhalo disruption. This study reshapes our understanding of the MW's impact on its satellites, suggesting it strips subhalos more efficiently than typical MW-mass galaxies due to its larger disk-to-halo mass ratio and earlier disk formation., Comment: 30 pages, 13 figures, 1 table. Submitted to ApJ, comments welcome

The video topic segmentation (VTS) task segments videos into intelligible, non-overlapping topics, facilitating efficient comprehension of video content and quick access to specific content. VTS is also critical to various downstream video understanding tasks. Traditional VTS methods using shallow features or unsupervised approaches struggle to accurately discern the nuances of topical transitions. Recently, supervised approaches have achieved superior performance on video action or scene segmentation over unsupervised approaches. In this work, we improve supervised VTS by thoroughly exploring multimodal fusion and multimodal coherence modeling. Specifically, (1) we enhance multimodal fusion by exploring different architectures using cross-attention and mixture of experts. (2) To generally strengthen multimodality alignment and fusion, we pre-train and fine-tune the model with multimodal contrastive learning. (3) We propose a new pre-training task tailored for the VTS task, and a novel fine-tuning task for enhancing multimodal coherence modeling for VTS. We evaluate the proposed approaches on educational videos, in the form of lectures, due to the vital role of topic segmentation of educational videos in boosting learning experiences. Additionally, we introduce a large-scale Chinese lecture video dataset to augment the existing English corpus, promoting further research in VTS. Experiments on both English and Chinese lecture datasets demonstrate that our model achieves superior VTS performance compared to competitive unsupervised and supervised baselines.

We report on high-sensitivity neutral hydrogen observations toward the gas-rich interacting galaxies NGC 3395/3396 with the Five-hundred-meter Aperture Spherical radio Telescope (FAST). Compared to previous observations carried out by the Very Large Array (VLA) and the Westerbork Synthesis Radio Telescope (WSRT), a more extended HI envelope around this system has been detected. The total HI gas mass of the NGC 3395/3396 system is estimated to be 7.8 x 109 M. This value is 2.7 times more than that reported based on the VLA interferometric maps. Previous observations found a large HI tail extending to the south-west and a minor tail emerging from the north of this peculiar galaxy pair. Based on the high-sensitivity observations of FAST, an extended HI plume to the north-west and a gas plume to the north-east have been detected for the first time. Neutral hydrogen of the two smaller galaxies IC 2604 and IC 2608 on the south of the system have also been detected. We discuss the origins of these extra gas and possible tidal interactions between these galaxies. NGC 3395/3396's most prominent tidal feature, the south-west tail combined with the new detected north-west plume behaves like a large ring. We suggest the ring might be formed by the previous fly-by interaction between NGC 3395 and NGC 3396 which happened 500 Myr ago. Our study shows that high-sensitivity HI observations are important in revealing low column density gas, which is crucial to a deeper understanding of this interacting system.

Superconducting qubits, a promising candidate for universal quantum computing, currently face limitations in chip size due to reproducibility, wiring complexity, and packaging modes. Distributed quantum modules offer a viable strategy for constructing larger quantum information processing systems, though universal quantum gate operations between remote qubits have yet to be realized. Here, we demonstrate high-fidelity perfect entanglers between two remote superconducting quantum devices over 30 cm distance, leveraging the standing-wave modes in the coaxial cable connecting them. We achieve cross-entropy benchmarking (XEB) fidelities of $(99.15 \pm 0.02)\%$ and $(98.04 \pm 0.04)\%$ for CNOT and CZ gates, respectively, which are more efficient and universal than existing state transfer or feedback-based protocols. This advancement significantly enhances the feasibility of universal distributed quantum information processing, essential for the future development of large-scale quantum systems.

Dynamically localized states in quantum many-body systems are fundamentally important in understanding quantum thermalization and have applications in quantum information processing. Here we explore many-body dynamical localization (MBDL) without disorders in a non-integrable quantum XY spin chain under periodical and quadratic kicks. We obtain the localization phase diagram with the MBDL and delocalization states and show dynamical observables to extract the phase diagram. For proper kick strengths in the MBDL regime, we reveal a local dynamical decoupling effect for persistent Rabi oscillation of certain spins. Furthermore, we propose the MBDL-protected quantum information at high temperatures, and present an analysis of the dynamical decoupling to obtain the required system parameters for quantum storage. Compared to other non-thermalized states, the disorder-free MBDL states require much fewer repetitions and resources, providing a promising way to protect and store quantum information robust against thermal noises., Comment: 6+4 pages, 4+6 figures; some references are added

The rotation curves of spiral galaxies exhibit a great diversity that challenge our understanding of galaxy formation and the nature of dark matter. Previous studies showed that in self-interacting dark matter (SIDM) models with a cross section per unit mass of $\sigma/m\approx{\cal O}(1)~{\rm cm^2/g}$, the predicted dark matter central densities are a good match to the observed densities in galaxies. In this work, we explore a regime with a larger cross section of $\sigma/m\approx20-40~{\rm cm^2/g}$ in dwarf galactic halos. We will show that such strong dark matter self-interactions can further amplify the diversity of halo densities inherited from their assembly history. High concentration halos can enter the gravothermal collapse phase within $10~{\rm Gyr}$, resulting in a high density, while low concentration ones remain in the expansion phase and have a low density. We fit the rotation curves of $14$ representative low surface brightness galaxies and demonstrate how the large range of observed central densities are naturally accommodated in the strong SIDM regime of $\sigma/m\approx20-40~{\rm cm^2/g}$. Galaxies that are outliers in the previous studies due to their high halo central densities, are no longer outliers in this SIDM regime as their halos would be in the collapse phase. For galaxies with a low density, the SIDM fits are robust to the variation of the cross section. Our findings open up a new window for testing gravothermal collapse, the unique signature of strong dark matter self-interactions, and exploring broad SIDM model space., Comment: 21 pages, 14 figures

Cavity expansion is a canonical problem in geotechnics, which can be described by partial differential equations (PDEs) and ordinary differential equations (ODEs). This study explores the potential of using a new solver, a physics-informed neural network (PINN), to calculate the stress field in an expanded cavity in the elastic and elasto-plastic regimes. Whilst PINNs have emerged as an effective universal function approximator for deriving the solutions of a wide range of governing PDEs/ODEs, their ability to solve elasto-plastic problems remains uncertain. A novel parsimonious loss function is first proposed to balance the simplicity and accuracy of PINN. The proposed method is applied to diverse material behaviours in the cavity expansion problem including isotropic, anisotropic elastic media, and elastic-perfectly plastic media with Tresca and Mohr-Coulomb yield criteria. The results indicate that the use of a parsimonious prior information-based loss function is highly beneficial to deriving the approximate solutions of complex PDEs with high accuracy. The present method allows for accurate derivation of solutions for both elastic and plastic mechanical responses of an expanded cavity. It also provides insights into how PINNs can be further advanced to solve more complex problems in geotechnical practice.

The Transformer architecture has significantly advanced deep learning, particularly in natural language processing, by effectively managing long-range dependencies. However, as the demand for understanding complex relationships grows, refining the Transformer's architecture becomes critical. This paper introduces Skip-Layer Attention (SLA) to enhance Transformer models by enabling direct attention between non-adjacent layers. This method improves the model's ability to capture dependencies between high-level abstract features and low-level details. By facilitating direct attention between these diverse feature levels, our approach overcomes the limitations of current Transformers, which often rely on suboptimal intra-layer attention. Our implementation extends the Transformer's functionality by enabling queries in a given layer to interact with keys and values from both the current layer and one preceding layer, thus enhancing the diversity of multi-head attention without additional computational burden. Extensive experiments demonstrate that our enhanced Transformer model achieves superior performance in language modeling tasks, highlighting the effectiveness of our skip-layer attention mechanism., Comment: 7 pages, 1 figure

We systemically evaluate the performance of the self-interacting dark matter (SIDM) halo model proposed in arXiv:2305.16176 with matched halos from high-resolution cosmological CDM and SIDM simulations. The model incorporates SIDM effects along mass evolution histories of CDM halos and it is applicable to both isolated halos and suhbhalos. We focus on the accuracy of the model in predicting halo density profiles at $z=0$ and the evolution of maximum circular velocity. We find the model predictions agree with the simulations within $10\%-50\%$ for most of the simulated (sub)halos, $50\%-100\%$ for extreme cases. This indicates that the model effectively captures the gravothermal evolution of the halos with very strong, velocity-dependent self-interactions. For an example application, we apply the model to study the impact of various SIDM scenarios on strong lensing perturber systems, demonstrating its utility in predicting SIDM effects for small-scale structure analyses. Our findings confirm that the model is an effective tool for mapping CDM halos into their SIDM counterparts., Comment: 20 pages, 19 figures

Named Entity Recognition and Relation Extraction are two crucial and challenging subtasks in the field of Information Extraction. Despite the successes achieved by the traditional approaches, fundamental research questions remain open. First, most recent studies use parameter sharing for a single subtask or shared features for both two subtasks, ignoring their semantic differences. Second, information interaction mainly focuses on the two subtasks, leaving the fine-grained informtion interaction among the subtask-specific features of encoding subjects, relations, and objects unexplored. Motivated by the aforementioned limitations, we propose a novel model to jointly extract entities and relations. The main novelties are as follows: (1) We propose to decouple the feature encoding process into three parts, namely encoding subjects, encoding objects, and encoding relations. Thanks to this, we are able to use fine-grained subtask-specific features. (2) We propose novel inter-aggregation and intra-aggregation strategies to enhance the information interaction and construct individual fine-grained subtask-specific features, respectively. The experimental results demonstrate that our model outperforms several previous state-of-the-art models. Extensive additional experiments further confirm the effectiveness of our model.

We combine the semi-analytical structure formation model, SASHIMI, which predicts subhalo populations in collisionless, cold dark matter (CDM), with a parametric model that maps CDM halos to self-interacting dark matter (SIDM) halos. The resulting model, SASHIMI-SIDM, generates SIDM subhalo populations down to sub-galactic mass scales, for an arbitrary input cross section, in minutes. We show that SASHIMI-SIDM agrees with SIDM subhalo populations from high-resolution cosmological zoom-in simulations in resolved regimes. Crucially, we predict that the fraction of core-collapsed subhalos peaks at a mass scale determined by the input SIDM cross section and decreases toward higher halo masses, consistent with the predictions of gravothermal models and cosmological simulations. For the first time, we also show that the core-collapsed fraction decreases toward lower halo masses; this result is uniquely enabled by our semi-analytical approach. As a proof of principle, we apply SASHIMI-SIDM to predict the boost to the local dark matter density and annihilation rate from core-collapsed SIDM subhalos, which can be enhanced relative to CDM by an order of magnitude for viable SIDM models. Thus, SASHIMI-SIDM provides an efficient and reliable tool for scanning SIDM parameter space and testing it with astrophysical observations. The code is publicly available at https://github.com/shinichiroando/sashimi-si., Comment: 21 pages, 7 figures. The code accompanying this submission is available at https://github.com/shinichiroando/sashimi-si

Relaxation processes are crucial in understanding the structural rearrangements of liquids and amorphous materials. However, the overarching principle that governs these processes across vastly different materials remains an open question. Substantial analysis has been carried out based on the motions of individual particles. Here, alternatively, we propose viewing the global configuration as a single entity. We introduce a global order parameter, namely the inherent structure minimal displacement (IS Dmin), to quantify the variability of configurations by a pattern-matching technique. Through atomic simulations of seven model glass-forming liquids, we unify the influences of temperature, pressure, and perturbation time on the relaxation dissipation, via a scaling law between the mechanical damping factor and IS Dmin. Fundamentally, this scaling reflects the curvature of the local potential energy landscape. Our findings uncover a universal origin of glassy relaxation and offer an alternative approach to studying disordered systems., Comment: 22 pages, 6 figures

Dark matter (DM) halos can be subject to gravothermal collapse if the DM is not collisionless, but engaged in strong self-interactions. When the scattering can efficiently transfer heat from the centre to the outskirts, the central region of the halo collapses and reaches densities much higher than those for collisionless DM. This phenomenon is potentially observable in studies of strong lensing. Current theoretical efforts are motivated by observations of surprisingly dense substructures. A comparison with observations requires accurate predictions. One method to obtain such predictions is to use N-body simulations. Collapsed halos are extreme systems that pose challenges when applying state-of-the-art codes to model self-interacting dark matter (SIDM). We investigate the root of such problems, with a focus on energy non-conservation and discuss possible strategies to avoid them. We ran N-body simulations, with and without SIDM, of an isolated DM-only halo and we adjusted the numerical parameters to check the accuracy of the simulation. We find that not only the numerical scheme for SIDM can lead to energy non-conservation, but also the modelling of gravitational interaction and the time integration are problematic. The issues we find are: (a) particles changing their time step in a non-time-reversible manner; (b) the asymmetry in the tree-based gravitational force evaluation; and (c) SIDM velocity kicks breaking the time symmetry. Tuning the parameters of the simulation allows us to conserve energy not only at early stages of the evolution, but also later on. However, the cost of the simulations becomes prohibitively large. Some of the problems that make the simulations of the gravothermal collapse phase inaccurate can be overcome by choosing appropriate numerical schemes. However, other issues still pose a challenge. Our findings motivate further works on addressing these challenges., Comment: 11 pages, 4 figures

van Dokkum et al. (arXiv:2309.07969) reported the discovery of JWST-ER1, a strong lensing object at redshift $z\approx2$, using data from the James Webb Space Telescope. The lens mass within the Einstein ring is $5.9$ times higher than the expected stellar mass from a Chabrier initial mass function, indicating a high dark matter density. In this work, we show that a cold dark matter halo, influenced by gas-driven adiabatic contraction, can account for the observed lens mass. We interpret the measurement of JWST-ER1 in the self-interacting dark matter scenario and show that the cross section per particle mass $\sigma/m\approx0.1~{\rm cm^2/g}$ is generally favored. Intriguingly, $\sigma/m\approx0.1~{\rm cm^2/g}$ can also be consistent with the strong lensing observations of early-type galaxies at redshift $z\approx0.2$, where adiabatic contraction is not observed overall., Comment: 11 pages, 7 figures, Accepted for publication in ApJL

In perception, multiple sensory information is integrated to map visual information from 2D views onto 3D objects, which is beneficial for understanding in 3D environments. But in terms of a single 2D view rendered from different angles, only limited partial information can be provided.The richness and value of Multi-view 2D information can provide superior self-supervised signals for 3D objects. In this paper, we propose a novel self-supervised point cloud representation learning method, MM-Point, which is driven by intra-modal and inter-modal similarity objectives. The core of MM-Point lies in the Multi-modal interaction and transmission between 3D objects and multiple 2D views at the same time. In order to more effectively simultaneously perform the consistent cross-modal objective of 2D multi-view information based on contrastive learning, we further propose Multi-MLP and Multi-level Augmentation strategies. Through carefully designed transformation strategies, we further learn Multi-level invariance in 2D Multi-views. MM-Point demonstrates state-of-the-art (SOTA) performance in various downstream tasks. For instance, it achieves a peak accuracy of 92.4% on the synthetic dataset ModelNet40, and a top accuracy of 87.8% on the real-world dataset ScanObjectNN, comparable to fully supervised methods. Additionally, we demonstrate its effectiveness in tasks such as few-shot classification, 3D part segmentation and 3D semantic segmentation., Comment: Accepted by AAAI 2024

Correlated errors can significantly impact the quantum error correction, which challenges the assumption that errors occur in different qubits independently in both space and time. Superconducting qubits have been found to suffer correlated errors across multiple qubits, which could be attributable to ionizing radiations and cosmic rays. Nevertheless, the direct evidence and a quantitative understanding of this relationship are currently lacking. In this work, we propose to continuously monitor multi-qubit simultaneous charge-parity jumps to detect correlated errors and find that occur more frequently than multi-qubit simultaneous bit flips. Then, we propose to position two cosmic-ray muon detectors directly beneath the sample box in a dilution refrigerator and successfully observe the correlated errors in a superconducting qubit array triggered by muons. By introducing a lead shielding layer on the refrigerator, we also reveal that the majority of other correlated errors are primarily induced by gamma rays. Furthermore, we find the superconducting film with a higher recombination rate of quasiparticles used in the qubits is helpful in reducing the duration of correlated errors. Our results provide experimental evidence of the impact of gamma rays and muons on superconducting quantum computation and offer practical insights into mitigation strategies for quantum error correction. In addition, we observe the average occurrence rate of muon-induced correlated errors in our processor is approximately 0.40 min$^{-1}$cm$^{-2}$, which is comparable to the muon event rate detected by the muon detector with 0.506 min$^{-1}$cm$^{-2}$. This demonstrates the potential applications of superconducting qubit arrays as low-energy threshold sensors in the field of high-energy physics., Comment: 7 pages and 5 figures for the main text, 20 pages and 20 figures for the supplementary materials

The satellite galaxy Crater II of the Milky Way is extremely cold and exceptionally diffuse. These unusual properties are challenging to understand in the standard model of cold dark matter. We investigate the formation of Crater II in self-interacting dark matter (SIDM), where dark matter particles can scatter and thermalize. We conduct a series of controlled N-body simulations to model the tidal evolution of Crater II, varying the self-interacting cross section, orbit parameters, and initial stellar distribution. Dark matter self-interactions lead to halo core formation and the distribution of stars expands accordingly. A cored SIDM halo also boosts tidal mass loss, allowing for a high orbit. Our simulations show that SIDM halos with a $1~{\rm kpc}$ core can simultaneously explain the low stellar velocity dispersion and the large half-light radius of Crater II, remaining robust to the initial distribution of stars. For the orbit motivated by the measurements from Gaia Early Data Release 3, the favored self-interacting cross section is approximately $60~{\rm cm^2/g}$ on the mass scale of Crater II., Comment: 9 pages, 5 figures; Revised for publication in The Astrophysical Journal Letters

Recently, unified speech-text models, such as SpeechGPT, VioLA, and AudioPaLM, have achieved remarkable performance on various speech tasks. These models discretize speech signals into tokens (speech discretization) and use a shared vocabulary for both text and speech tokens. Then they train a single decoder-only Transformer on a mixture of speech tasks. However, these models rely on the Loss Masking strategy for the ASR task, which ignores the dependency among speech tokens. In this paper, we propose to model speech tokens in an autoregressive way, similar to text. We find that applying the conventional cross-entropy loss on input speech tokens does not consistently improve the ASR performance over the Loss Masking approach. To address this issue, we propose a novel approach denoted Smoothed Label Distillation (SLD), which applies a KL divergence loss with smoothed labels on speech tokens. Our experiments show that SLD effectively models speech tokens and outperforms Loss Masking for decoder-only Transformers in ASR tasks with different speech discretization methods. The source code can be found here: https://github.com/alibaba-damo-academy/SpokenNLP/tree/main/sld, Comment: 5 pages, accepted by ICASSP 2024

Topic segmentation is critical for obtaining structured documents and improving downstream tasks such as information retrieval. Due to its ability of automatically exploring clues of topic shift from abundant labeled data, recent supervised neural models have greatly promoted the development of long document topic segmentation, but leaving the deeper relationship between coherence and topic segmentation underexplored. Therefore, this paper enhances the ability of supervised models to capture coherence from both logical structure and semantic similarity perspectives to further improve the topic segmentation performance, proposing Topic-aware Sentence Structure Prediction (TSSP) and Contrastive Semantic Similarity Learning (CSSL). Specifically, the TSSP task is proposed to force the model to comprehend structural information by learning the original relations between adjacent sentences in a disarrayed document, which is constructed by jointly disrupting the original document at topic and sentence levels. Moreover, we utilize inter- and intra-topic information to construct contrastive samples and design the CSSL objective to ensure that the sentences representations in the same topic have higher similarity, while those in different topics are less similar. Extensive experiments show that the Longformer with our approach significantly outperforms old state-of-the-art (SOTA) methods. Our approach improve $F_1$ of old SOTA by 3.42 (73.74 -> 77.16) and reduces $P_k$ by 1.11 points (15.0 -> 13.89) on WIKI-727K and achieves an average relative reduction of 4.3% on $P_k$ on WikiSection. The average relative $P_k$ drop of 8.38% on two out-of-domain datasets also demonstrates the robustness of our approach., Comment: Accepted by EMNLP 2023. Codes is available at https://github.com/alibaba-damo-academy/SpokenNLP/

Ever since the launch of ChatGPT in 2022, a rising concern is whether ChatGPT will replace programmers and kill jobs. Motivated by this widespread concern, we conducted an empirical study to systematically compare ChatGPT against programmers in question-answering and software-maintaining. We reused a dataset introduced by prior work, which includes 130 StackOverflow (SO) discussion threads referred to by the Java developers of 357 GitHub projects. We mainly investigated three research questions (RQs). First, how does ChatGPT compare with programmers when answering technical questions? Second, how do developers perceive the differences between ChatGPT's answers and SO answers? Third, how does ChatGPT compare with humans when revising code for maintenance requests? For RQ1, we provided the 130 SO questions to ChatGPT, and manually compared ChatGPT answers with the accepted/most popular SO answers in terms of relevance, readability, informativeness, comprehensiveness, and reusability. For RQ2, we conducted a user study with 30 developers, asking each developer to assess and compare 10 pairs of answers, without knowing the information source (i.e., ChatGPT or SO). For RQ3, we distilled 48 software maintenance tasks from 48 GitHub projects citing the studied SO threads. We queried ChatGPT to revise a given Java file, and to incorporate the code implementation for any prescribed maintenance requirement. Our study reveals interesting phenomena: For the majority of SO questions (97/130), ChatGPT provided better answers; in 203 of 300 ratings, developers preferred ChatGPT answers to SO answers; ChatGPT revised code correctly for 22 of the 48 tasks. Our research will expand people's knowledge of ChatGPT capabilities, and shed light on future adoption of ChatGPT by the software industry.