Search

Image Captioning is an important Language and Vision task that finds application in a variety of contexts, ranging from healthcare to autonomous vehicles. As many real-world applications rely on devices with limited resources, much effort in the field was put into the development of lighter and faster models. However, much of the current optimizations focus on the Transformer architecture in contrast to the existence of more efficient methods. In this work, we introduce SwiFTeR, an architecture almost entirely based on Fourier Transform and Retention, to tackle the main efficiency bottlenecks of current light image captioning models, being the visual backbone's onerosity, and the decoder's quadratic cost. SwiFTeR is made of only 20M parameters, and requires 3.1 GFLOPs for a single forward pass. Additionally, it showcases superior scalability to the caption length and its small memory requirements enable more images to be processed in parallel, compared to the traditional transformer-based architectures. For instance, it can generate 400 captions in one second. Although, for the time being, the caption quality is lower (110.2 CIDEr-D), most of the decrease is not attributed to the architecture but rather an incomplete training practice which currently leaves much room for improvements. Overall, SwiFTeR points toward a promising direction to new efficient architectural design. The implementation code will be released in the future., Comment: Pre-print version of paper accepted for ICONIP 2024

Autoregressive Sequence-To-Sequence models are the foundation of many Deep Learning achievements in major research fields such as Vision and Natural Language Processing. Despite that, they still present significant limitations. For instance, when errors occur in the early steps of the prediction, the whole output is severely affected. Such reliance on previously predicted tokens and the inherent computational unfriendliness of sequential algorithms, motivated researchers to explore different architectures and methods in the search for bidirectional approaches. In this work, we introduce the Bidirectional Awareness Induction (BAI), a training method that leverages a subset of elements in the network, the Pivots, to perform bidirectional learning without breaking the autoregressive constraints. To showcase its flexibility, we apply the method to three architectures, the Transformer, ExpansionNet v2 and GPT, then perform experiments over three tasks. Experimental results showcase BAI's effectiveness on all selected tasks and architectures. In particular, we observed an increase of up to 2.4 CIDEr in Image-Captioning, 4.96 BLEU in Neural Machine Translation, and 1.16 ROUGE in Text Summarization compared to the respective baselines. Notably, BAI not only has a positive impact on models trained from scratch but on pre-trained models as well. Such an aspect, combined with the absence of architectural requirements synergizes well with the current trend of LLMs.

We theoretically demonstrate that carbon nanoscrolls -- spirally wrapped graphene layers with open endpoints -- can be characterized by a large positive magnetoconductance. We show that when a carbon nanoscroll is subject to an axial magnetic field of ~ 10T, the ballistic conductance at low carrier densities of the nanoscroll has an increase of about 200%. Importantly, we find that this positive magnetoconductance is not only preserved but can be even enhanced in the presence of on-site disorder. We prove that the positive magnetoconductance comes about the emergence of magnetic field-induced zero energy modes, specific of rolled-up geometries. Our results establish curved graphene systems as a new material platform displaying sizable magnetoresistive phenomena., Comment: 13 pages, 4 figures

The gas-phase metallicity of galaxies is regulated by multiple astrophysical processes, which makes it a crucial diagnostic of galaxy formation and evolution. Beyond the fundamental mass-metallicity relation, a debate about the secondary galaxy property to predict the metallicity of galaxies arises. Motivated by this, we systematically examine the relationship between gas-phase metallicity and other galaxy properties, i.e. star formation rate (SFR) and galaxy size, in addition to stellar mass in both observation and simulation. We utilize the data from the MaNGA (Mapping Nearby Galaxies at Apache Point Observatory) survey and the TNG50 simulations. We find that the combination of $M_*/R_{\rm e}^\beta$ with $\beta\sim 0.6-1$ is in much stronger correlation to the metallicity than stellar mass alone, regardless of whether the SFR is included or not, in both observation and simulation. This indicates that galaxy size plays a more important role in determining gas-phase metallicity of galaxies than SFR. In addition, the TNG simulation predicts that the SFR, although being a subdominant role, becomes increasingly important in high-$z$ universe. Finally, we speculate that SFR modulates metallicity on the temporal dimension, synchronized with time-varying gas inflows, and galaxy size regulates metallicity on the spatial dimension by affecting the gravitational potential and the mass loading factor., Comment: 11 pages, 6 figures, ApJL accepted

The motion of the barycenter of the Solar System (SSB), the origin of the International Celestial Reference System, causes a directional displacement known as secular aberration. The secular aberration drift caused by the galactocentric acceleration of the SSB has been modeled in the third generation of the International Celestial Reference Frame. We aim to address another secular aberration drift effect due to the change in the line-of-sight direction and study its implications for stellar proper motions. A complete formula of secular aberration drift is derived, and its influence on stellar proper motion is computed based on the astrometric data in \textit{Gaia} Data Release 3. We found that the secular aberration drift due to the change in the line-of-sight direction tends to decrease the observed proper motions for stars with galactic longitudes between $0^{\circ}$ and $180^{\circ}$, and increase the observed proper motion for stars in the remaining region. If this secular aberration drift effect is ignored, it will induce an additional proper motion of $>1\,\mathrm{mas\,yr^{-1}}$ for 84 stars and $>0.02\,\mathrm{mas\,yr^{-1}}$ for 5\,944\,879 stars, which is comparable to or several times greater than the typical formal uncertainty of the \textit{Gaia} proper motion measurements at $G<13$. The secular aberration drift due to the change in the line-of-sight direction and the acceleration of the SSB should be modeled to make the stellar reference frame consistent with the extragalactic reference frame., Comment: 5 pages, 5 figures. Accepted for publication at A&A

We establish a new criterion for exponential mixing of random dynamical systems. Our criterion is applicable to a wide range of systems, including in particular dispersive equations. Its verification is in nature related to several topics, i.e., asymptotic compactness in dynamical systems, global stability of evolution equations, and localized control problems. As an initial application, we exploit the exponential mixing of random nonlinear wave equations with degenerate damping, critical nonlinearity, and physically localized noise. The essential challenge lies in the fact that the weak dissipation and randomness interact in the evolution.

The matrix product state (MPS) ansatz offers a promising approach for finding the ground state of molecular Hamiltonians and solving quantum chemistry problems. Building on this concept, the proposed technique of quantum circuit MPS (QCMPS) enables the simulation of chemical systems using a relatively small number of qubits. In this study, we enhance the optimization performance of the QCMPS ansatz by employing the variational quantum imaginary time evolution (VarQITE) approach. Guided by McLachlan's variational principle, the VarQITE method provides analytical metrics and gradients, resulting in improved convergence efficiency and robustness of the QCMPS. We validate these improvements numerically through simulations of $\rm H_2$, $\rm H_4$, and $\rm LiH$ molecules. Additionally, given that VarQITE is applicable to non-Hermitian Hamiltonians, we evaluate its effectiveness in preparing the ground state of transcorrelated (TC) Hamiltonians. This approach yields energy estimates comparable to the complete basis set (CBS) limit while using even fewer qubits. Specifically, we perform simulations of the beryllium atom and $\rm LiH$ molecule using only three qubits, maintaining high fidelity with the CBS ground state energy of these systems. This qubit reduction is achieved through the combined advantages of both the QCMPS ansatz and transcorrelation. Our findings demonstrate the potential practicality of this quantum chemistry algorithm on near-term quantum devices., Comment: 15 pages, 8 figures

Multi Object Tracking (MOT) in ice hockey pursues the combined task of localizing and associating players across a given sequence to maintain their identities. Tracking players from monocular broadcast feeds is an important computer vision problem offering various downstream analytics and enhanced viewership experience. However, existing trackers encounter significant difficulties in dealing with occlusions, blurs, and agile player movements prevalent in telecast feeds. In this work, we propose a novel tracking approach by formulating MOT as a bipartite graph matching problem infused with homography. We disentangle the positional representations of occluded and overlapping players in broadcast view, by mapping their foot keypoints to an overhead rink template, and encode these projected positions into the graph network. This ensures reliable spatial context for consistent player tracking and unfragmented tracklet prediction. Our results show considerable improvements in both the IDsw and IDF1 metrics on the two available broadcast ice hockey datasets., Comment: Accepted at the Conference on Robots and Vision (CRV), 2024

The advent of Neural-network Quantum States (NQS) has significantly advanced wave function ansatz research, sparking a resurgence in orbital space variational Monte Carlo (VMC) exploration. This work introduces three algorithmic enhancements to reduce computational demands of VMC optimization using NQS: an adaptive learning rate algorithm, constrained optimization, and block optimization. We evaluate the refined algorithm on complex multireference bond stretches of $\rm H_2O$ and $\rm N_2$ within the cc-pVDZ basis set and calculate the ground-state energy of the strongly correlated chromium dimer ($\rm Cr_2$) in the Ahlrichs SV basis set. Our results achieve superior accuracy compared to coupled cluster theory at a relatively modest CPU cost. This work demonstrates how to enhance optimization efficiency and robustness using these strategies, opening a new path to optimize large-scale Restricted Boltzmann Machine (RBM)-based NQS more effectively and marking a substantial advancement in NQS's practical quantum chemistry applications., Comment: 13 pages, 9 figures, and 2 tables

Optical vortex beam with orbital angular momentum (OAM) has great potential to increase the capacity of optical communication and information processing in classical and quantum regimes. Nevertheless, important challenges that influence the optical data transmission in free space is the existence of diffusers along the optical path, which causes inevitable information loss during the wave propagation. Numerous algorithms have been proposed successively for identifying the modes of vortex beams propagating through scattering media. However, these methods all require completion on a computer, which is energyintensive and energy consuming. Here, we propose an all-optical regime for identifying the modes of vortex light fields propagating through scattering media. After training by deep learning, our model can recognize the mode of vortex beam through unknown phase diffusers, demonstrating generalization to new random diffusers that have never been encountered before. Once physically deployed, the entire setup will rapidly identify the modes of vortex light propagating through scattering media at the speed of light, and the entire inference process will consume zero energy except for illumination source. Our research represents a significant step towards highly accurate recognition of vortex light modes propagating through complex scattering media, providing significant guidance for the application of optical communication in complex environments.

High-order topological phases of matter refer to the systems of $n$-dimensional bulk with the topology of $m$-th order, exhibiting $(n-m)$-dimensional boundary modes and can be characterized by topological pumping. Here, we experimentally demonstrate two types of second-order topological pumps, forming four 0-dimensional corner localized states on a 4$\times$4 square lattice array of 16 superconducting qubits. The initial ground state of the system for half-filling, as a product of four identical entangled 4-qubit states, is prepared using an adiabatic scheme. During the pumping procedure, we adiabatically modulate the superlattice Bose-Hubbard Hamiltonian by precisely controlling both the hopping strengths and on-site potentials. At the half pumping period, the system evolves to a corner-localized state in a quadrupole configuration. The robustness of the second-order topological pump is also investigated by introducing different on-site disorder. Our work studies the topological properties of high-order topological phases from the dynamical transport picture using superconducting qubits, which would inspire further research on high-order topological phases.

Although the Transformer is currently the best-performing architecture in the homogeneous configuration (self-attention only) in Neural Machine Translation, many State-of-the-Art models in Natural Language Processing are made of a combination of different Deep Learning approaches. However, these models often focus on combining a couple of techniques only and it is unclear why some methods are chosen over others. In this work, we investigate the effectiveness of integrating an increasing number of heterogeneous methods. Based on a simple combination strategy and performance-driven synergy criteria, we designed the Multi-Encoder Transformer, which consists of up to five diverse encoders. Results showcased that our approach can improve the quality of the translation across a variety of languages and dataset sizes and it is particularly effective in low-resource languages where we observed a maximum increase of 7.16 BLEU compared to the single-encoder model.

We use the density matrix renormalization group (DMRG) method to study the SU(4) $t$-$J$ chain. We find that, in addition to the conventional repulsive Luttinger liquid phase and phase separation, there are two phases in the attractive Luttinger liquid region dependent on whether the flavor gap is opened or not. The first with the flavor gap is the molecular superfluid phase (the SU(4) singlet instability) which is well-known in the attractive SU(4) Hubbard model ($U<0$). The second without the flavor gap is the superconducting phase (the six-component pairing instability). Furthermore, the molecular superfluid instability cannot coexist with the superconducting instability. This is general in SU($N$) models with $N>2$ and is well demonstrated by the theoretical analysis based on the phenomenological bosonization results., Comment: 11 pages, 6 figures

Characterizing the nature of hydrodynamical transport properties in quantum dynamics provides valuable insights into the fundamental understanding of exotic non-equilibrium phases of matter. Experimentally simulating infinite-temperature transport on large-scale complex quantum systems is of considerable interest. Here, using a controllable and coherent superconducting quantum simulator, we experimentally realize the analog quantum circuit, which can efficiently prepare the Haar-random states, and probe spin transport at infinite temperature. We observe diffusive spin transport during the unitary evolution of the ladder-type quantum simulator with ergodic dynamics. Moreover, we explore the transport properties of the systems subjected to strong disorder or a tilted potential, revealing signatures of anomalous subdiffusion in accompany with the breakdown of thermalization. Our work demonstrates a scalable method of probing infinite-temperature spin transport on analog quantum simulators, which paves the way to study other intriguing out-of-equilibrium phenomena from the perspective of transport., Comment: Main text: 13 pages, 5 figures; Supplementary: 17 pages, 16 figures, 1 table

Hockey rink registration is a useful tool for aiding and automating sports analysis. When combined with player tracking, it can provide location information of players on the rink by estimating a homography matrix that can warp broadcast video frames onto an overhead template of the rink, or vice versa. However, most existing techniques require accurate ground truth information, which can take many hours to annotate, and only work on the trained rink types. In this paper, we propose a generalized rink registration pipeline that, once trained, can be applied to both seen and unseen rink types with only an overhead rink template and the video frame as inputs. Our pipeline uses domain adaptation techniques, semi-supervised learning, and synthetic data during training to achieve this ability and overcome the lack of non-NHL training data. The proposed method is evaluated on both NHL (source) and non-NHL (target) rink data and the results demonstrate that our approach can generalize to non-NHL rinks, while maintaining competitive performance on NHL rinks.

In this paper, we propose a novel fully Bayesian approach for the massive multiple-input multiple-output (MIMO) massive unsourced random access (URA). The payload of each user device is coded by the sparse regression codes (SPARCs) without redundant parity bits. A Bayesian model is established to capture the probabilistic characteristics of the overall system. Particularly, we adopt the core idea of the model-based learning approach to establish a flexible Bayesian channel model to adapt the complex environments. Different from the traditional divide-and-conquer or pilot-based massive MIMO URA strategies, we propose a three-layer message passing (TLMP) algorithm to jointly decode all the information blocks, as well as acquire the massive MIMO channel, which adopts the core idea of the variational message passing and approximate message passing. We verify that our proposed TLMP significantly enhances the spectral efficiency compared with the state-of-the-arts baselines, and is more robust to the possible codeword collisions.

In this paper, we design a novel two-phase unsourced random access (URA) scheme in massive multiple input multiple output (MIMO). In the first phase, we collect a sequence of information bits to jointly acquire the user channel state information (CSI) and the associated information bits. In the second phase, the residual information bits of all the users are partitioned into sub-blocks with a very short length to exhibit a higher spectral efficiency and a lower computational complexity than the existing transmission schemes in massive MIMO URA. By using the acquired CSI in the first phase, the sub-block recovery in the second phase is cast as a compressed sensing (CS) problem. From the perspective of the statistical physics, we provide a theoretical framework for our proposed URA scheme to analyze the induced problem based on the replica method. The analytical results show that the performance metrics of our URA scheme can be linked to the system parameters by a single-valued free entropy function. An AMP-based recovery algorithm is designed to achieve the performance indicated by the proposed theoretical framework. Simulations verify that our scheme outperforms the most recent counterparts., Comment: 16pages,7 figures

Neural-network quantum states (NQS) employ artificial neural networks to encode many-body wave functions in second quantization through variational Monte Carlo (VMC). They have recently been applied to accurately describe electronic wave functions of molecules and have shown the challenges in efficiency comparing with traditional quantum chemistry methods. Here we introduce a general non-stochastic optimization algorithm for NQS in chemical systems, which deterministically generates a selected set of important configurations simultaneously with energy evaluation of NQS. This method bypasses the need for Markov-chain Monte Carlo within the VMC framework, thereby accelerating the entire optimization process. Furthermore, this newly-developed non-stochastic optimization algorithm for NQS offers comparable or superior accuracy compared to its stochastic counterpart and ensures more stable convergence. The application of this model to test molecules exhibiting strong electron correlations provides further insight into the performance of NQS in chemical systems and opens avenues for future enhancements., Comment: 30 pages, 7 figures, and 1 table

The Image Captioning research field is currently compromised by the lack of transparency and awareness over the End-of-Sequence token () in the Self-Critical Sequence Training. If the token is omitted, a model can boost its performance up to +4.1 CIDEr-D using trivial sentence fragments. While this phenomenon poses an obstacle to a fair evaluation and comparison of established works, people involved in new projects are given the arduous choice between lower scores and unsatisfactory descriptions due to the competitive nature of the research. This work proposes to solve the problem by spreading awareness of the issue itself. In particular, we invite future works to share a simple and informative signature with the help of a library called SacreEOS. Code available at \emph{\href{https://github.com/jchenghu/sacreeos}{https://github.com/jchenghu/sacreeos}}

This study aims to investigate the systematics in planetary ephemeris reference frames through pulsar timing observations. We used the published data sets from several pulsar timing arrays and performed timing analyses for each pulsar using different planetary ephemerides retrieved from the Jet Propulsion Laboratory's Development Ephemeris (DE), Ephemeris of Planets and the Moon (EPM), and INPOP (Int\'egration Num\'erique Plan\'etaire de l'Observatoire de Paris). Then, we compared the timing solutions and modeled the differences in position and proper motion by vector spherical harmonics of the first degree. The timing solutions were also compared with those determined by very long baseline interferometry (VLBI) astrometry. The orientation offsets between the latest editions of the DE, EPM, and INPOP series do not exceed 0.4 milliarcseconds (mas), while the relative spins between these ephemerides are less than 5 microarcseconds per year ($\mathrm{\mu as\,yr^{-1}}$). We do not detect significant glides in either position or proper motion between these ephemerides. The orientation of the pulsar timing frames deviates from that of the VLBI frame from zero by approximately $\mathrm{0.4\,mas}$ when considering the formal uncertainty and possible systematics. The orientation of current planetary ephemeris frames is as accurate as at least 0.4 mas, and the nonrotating is better than $\mathrm{5\,\mu as\,yr^{-1}}$., Comment: 17 pages, 14 figures, 6 tables, to be accepted for publication at A&A