Search

Large Multimodal Models (LMMs) exhibit impressive cross-modal understanding and reasoning abilities, often assessed through multiple-choice questions (MCQs) that include an image, a question, and several options. However, many benchmarks used for such evaluations suffer from systematic biases. Remarkably, Large Language Models (LLMs) without any visual perception capabilities achieve non-trivial performance, undermining the credibility of these evaluations. To address this issue while maintaining the efficiency of MCQ evaluations, we propose MMEvalPro, a benchmark designed to avoid Type-I errors through a trilogy evaluation pipeline and more rigorous metrics. For each original question from existing benchmarks, human annotators augment it by creating one perception question and one knowledge anchor question through a meticulous annotation process. MMEvalPro comprises $2,138$ question triplets, totaling $6,414$ distinct questions. Two-thirds of these questions are manually labeled by human experts, while the rest are sourced from existing benchmarks (MMMU, ScienceQA, and MathVista). Compared with the existing benchmarks, our experiments with the latest LLMs and LMMs demonstrate that MMEvalPro is more challenging (the best LMM lags behind human performance by $31.73\%$, compared to an average gap of $8.03\%$ in previous benchmarks) and more trustworthy (the best LLM trails the best LMM by $23.09\%$, whereas the gap for previous benchmarks is just $14.64\%$). Our in-depth analysis explains the reason for the large performance gap and justifies the trustworthiness of evaluation, underscoring its significant potential for advancing future research., Comment: 21 pages, code released at https://github.com/chenllliang/MMEvalPro, Homepage at https://mmevalpro.github.io/

While quantum computing has strong potential in data-driven fields, the privacy issue of sensitive or valuable information involved in the quantum algorithm should be considered. Differential privacy (DP), which is a fundamental privacy tool widely used in the classical scenario, has been extended to the quantum domain, i.e., quantum differential privacy (QDP). QDP may become one of the most promising approaches toward privacy-preserving quantum computing since it is not only compatible with classical DP mechanisms but also achieves privacy protection by exploiting unavoidable quantum noise in noisy intermediate-scale quantum (NISQ) devices. This paper provides an overview of the various implementations of QDP and their performance in terms of privacy parameters under the DP setting. Specifically, we propose a taxonomy of QDP techniques, categorizing the literature on whether internal or external randomization is used as a source to achieve QDP and how these implementations are applied to each phase of the quantum algorithm. We also discuss challenges and future directions for QDP. By summarizing recent advancements, we hope to provide a comprehensive, up-to-date review for researchers venturing into this field., Comment: To be published in QCE24 (IEEE Quantum Week 2024)

This paper investigates traffic forecasting, which attempts to forecast the future state of traffic based on historical situations. This problem has received ever-increasing attention in various scenarios and facilitated the development of numerous downstream applications such as urban planning and transportation management. However, the efficacy of existing methods remains sub-optimal due to their tendency to model temporal and spatial relationships independently, thereby inadequately accounting for complex high-order interactions of both worlds. Moreover, the diversity of transitional patterns in traffic forecasting makes them challenging to capture for existing approaches, warranting a deeper exploration of their diversity. Toward this end, this paper proposes Conjoint Spatio-Temporal graph neural network (abbreviated as COOL), which models heterogeneous graphs from prior and posterior information to conjointly capture high-order spatio-temporal relationships. On the one hand, heterogeneous graphs connecting sequential observation are constructed to extract composite spatio-temporal relationships via prior message passing. On the other hand, we model dynamic relationships using constructed affinity and penalty graphs, which guide posterior message passing to incorporate complementary semantic information into node representations. Moreover, to capture diverse transitional properties to enhance traffic forecasting, we propose a conjoint self-attention decoder that models diverse temporal patterns from both multi-rank and multi-scale views. Experimental results on four popular benchmark datasets demonstrate that our proposed COOL provides state-of-the-art performance compared with the competitive baselines., Comment: Accepted by Information Fusion 2024

Intelligent Transportation System (ITS) is crucial for improving traffic congestion, reducing accidents, optimizing urban planning, and more. However, the complexity of traffic networks has rendered traditional machine learning and statistical methods less effective. With the advent of artificial intelligence, deep learning frameworks have achieved remarkable progress across various fields and are now considered highly effective in many areas. Since 2019, Graph Neural Networks (GNNs) have emerged as a particularly promising deep learning approach within the ITS domain, owing to their robust ability to model graph-structured data and address complex problems. Consequently, there has been increasing scholarly attention to the applications of GNNs in transportation, which have demonstrated excellent performance. Nevertheless, current research predominantly focuses on traffic forecasting, with other ITS domains, such as autonomous vehicles and demand prediction, receiving less attention. This paper aims to review the applications of GNNs across six representative and emerging ITS research areas: traffic forecasting, vehicle control system, traffic signal control, transportation safety, demand prediction, and parking management. We have examined a wide range of graph-related studies from 2018 to 2023, summarizing their methodologies, features, and contributions in detailed tables and lists. Additionally, we identify the challenges of applying GNNs in ITS and propose potential future research directions.

Quantum computing has been widely applied in various fields, such as quantum physics simulations, quantum machine learning, and big data analysis. However, in the domains of data-driven paradigm, how to ensure the privacy of the database is becoming a vital problem. For classical computing, we can incorporate the concept of differential privacy (DP) to meet the standard of privacy preservation by manually adding the noise. In the quantum computing scenario, researchers have extended classic DP to quantum differential privacy (QDP) by considering the quantum noise. In this paper, we propose a novel approach to satisfy the QDP definition by considering the errors generated by the projection operator measurement, which is denoted as shot noises. Then, we discuss the amount of privacy budget that can be achieved with shot noises, which serves as a metric for the level of privacy protection. Furthermore, we provide the QDP of shot noise in quantum circuits with depolarizing noise. Through numerical simulations, we show that shot noise can effectively provide privacy protection in quantum computing.

This paper studies the problem of traffic flow forecasting, which aims to predict future traffic conditions on the basis of road networks and traffic conditions in the past. The problem is typically solved by modeling complex spatio-temporal correlations in traffic data using spatio-temporal graph neural networks (GNNs). However, the performance of these methods is still far from satisfactory since GNNs usually have limited representation capacity when it comes to complex traffic networks. Graphs, by nature, fall short in capturing non-pairwise relations. Even worse, existing methods follow the paradigm of message passing that aggregates neighborhood information linearly, which fails to capture complicated spatio-temporal high-order interactions. To tackle these issues, in this paper, we propose a novel model named Dynamic Hypergraph Structure Learning (DyHSL) for traffic flow prediction. To learn non-pairwise relationships, our DyHSL extracts hypergraph structural information to model dynamics in the traffic networks, and updates each node representation by aggregating messages from its associated hyperedges. Additionally, to capture high-order spatio-temporal relations in the road network, we introduce an interactive graph convolution block, which further models the neighborhood interaction for each node. Finally, we integrate these two views into a holistic multi-scale correlation extraction module, which conducts temporal pooling with different scales to model different temporal patterns. Extensive experiments on four popular traffic benchmark datasets demonstrate the effectiveness of our proposed DyHSL compared with a broad range of competing baselines., Comment: Accepted by 2023 IEEE 39th International Conference on Data Engineering (ICDE 2023)

Node classification on graphs is a significant task with a wide range of applications, including social analysis and anomaly detection. Even though graph neural networks (GNNs) have produced promising results on this task, current techniques often presume that label information of nodes is accurate, which may not be the case in real-world applications. To tackle this issue, we investigate the problem of learning on graphs with label noise and develop a novel approach dubbed Consistent Graph Neural Network (CGNN) to solve it. Specifically, we employ graph contrastive learning as a regularization term, which promotes two views of augmented nodes to have consistent representations. Since this regularization term cannot utilize label information, it can enhance the robustness of node representations to label noise. Moreover, to detect noisy labels on the graph, we present a sample selection technique based on the homophily assumption, which identifies noisy nodes by measuring the consistency between the labels with their neighbors. Finally, we purify these confident noisy labels to permit efficient semantic graph learning. Extensive experiments on three well-known benchmark datasets demonstrate the superiority of our CGNN over competing approaches., Comment: Accepted by IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2023)

Graph neural networks have pushed state-of-the-arts in graph classifications recently. Typically, these methods are studied within the context of supervised end-to-end training, which necessities copious task-specific labels. However, in real-world circumstances, labeled data could be limited, and there could be a massive corpus of unlabeled data, even from unknown classes as a complementary. Towards this end, we study the problem of semi-supervised universal graph classification, which not only identifies graph samples which do not belong to known classes, but also classifies the remaining samples into their respective classes. This problem is challenging due to a severe lack of labels and potential class shifts. In this paper, we propose a novel graph neural network framework named UGNN, which makes the best of unlabeled data from the subgraph perspective. To tackle class shifts, we estimate the certainty of unlabeled graphs using multiple subgraphs, which facilities the discovery of unlabeled data from unknown categories. Moreover, we construct semantic prototypes in the embedding space for both known and unknown categories and utilize posterior prototype assignments inferred from the Sinkhorn-Knopp algorithm to learn from abundant unlabeled graphs across different subgraph views. Extensive experiments on six datasets verify the effectiveness of UGNN in different settings., Comment: Accepted by IEEE Transactions on Knowledge and Data Engineering (TKDE 2023)

Graph representation learning aims to effectively encode high-dimensional sparse graph-structured data into low-dimensional dense vectors, which is a fundamental task that has been widely studied in a range of fields, including machine learning and data mining. Classic graph embedding methods follow the basic idea that the embedding vectors of interconnected nodes in the graph can still maintain a relatively close distance, thereby preserving the structural information between the nodes in the graph. However, this is sub-optimal due to: (i) traditional methods have limited model capacity which limits the learning performance; (ii) existing techniques typically rely on unsupervised learning strategies and fail to couple with the latest learning paradigms; (iii) representation learning and downstream tasks are dependent on each other which should be jointly enhanced. With the remarkable success of deep learning, deep graph representation learning has shown great potential and advantages over shallow (traditional) methods, there exist a large number of deep graph representation learning techniques have been proposed in the past decade, especially graph neural networks. In this survey, we conduct a comprehensive survey on current deep graph representation learning algorithms by proposing a new taxonomy of existing state-of-the-art literature. Specifically, we systematically summarize the essential components of graph representation learning and categorize existing approaches by the ways of graph neural network architectures and the most recent advanced learning paradigms. Moreover, this survey also provides the practical and promising applications of deep graph representation learning. Last but not least, we state new perspectives and suggest challenging directions which deserve further investigations in the future., Comment: Accepted by Neural Networks 2024

Materials with AV$_3$Sb$_5$ (A=K, Rb, Cs) stoichiometry are recently discovered kagome superconductors with the electronic structure featuring a Dirac band, van Hove singularities and flat bands. These systems undergo anomalous charge-density-wave (CDW) transitions at $T_{\rm CDW}$~80-100 K, resulting in the reconstruction of the Fermi surface from the pristine phase. Although comprehensive investigations of the electronic structure via quantum oscillations (QOs) have been performed on the sister compounds CsV$_3$Sb$_5$ and RbV$_3$Sb$_5$, a detailed QO study of KV$_3$Sb$_5$ is so far absent. Here, we report the Shubnikov-de Haas QO study in KV$_3$Sb$_5$. We resolve a large number of new frequencies with the highest frequency of 2202 T (occupying ~54% of the Brillouin zone area in the $k_x$-$k_y$ plane). The Lifshitz-Kosevich analysis further gives relatively small cyclotron effective masses, and the angular dependence study reveals the two-dimensional nature of the frequencies with a sufficient signal-to-noise ratio. Finally, we compare the QO spectra for all three AV$_3$Sb$_5$ compounds collected under the same conditions, enabling us to point out the similarities and differences across these systems. Our results fill in the gap of the QO study in KV$_3$Sb$_5$ and provide valuable data to understand the band structure of all three members of AV$_3$Sb$_5$., Comment: 8 pages, 4 figures

The kagome metal CsV$_{3}$Sb$_{5}$ features an unusual competition between the charge-density-wave (CDW) order and superconductivity. Evidence for time-reversal symmetry breaking (TRSB) inside the CDW phase has been accumulating. Hence, the superconductivity in CsV$_{3}$Sb$_{5}$ emerges from a TRSB normal state, potentially resulting in an exotic superconducting state. To reveal the pairing symmetry, we first investigate the effect of nonmagnetic impurity. Our results show that the superconducting critical temperature is insensitive to disorder, pointing to conventional $s$-wave superconductivity. Moreover, our measurements of the self-field critical current ($I_{c,sf}$), which is related to the London penetration depth, also confirm conventional $s$-wave superconductivity with strong coupling. Finally, we measure $I_{c,sf}$ where the CDW order is removed by pressure and superconductivity emerges from the pristine normal state. Our results show that $s$-wave gap symmetry is retained, providing strong evidence for the presence of conventional $s$-wave superconductivity in CsV$_{3}$Sb$_{5}$ irrespective of the presence of the TRSB, Comment: 8 pages, 4 figures. Nano Letters (in press)

AV$_3$Sb$_5$ (A=Cs, K, Rb) are recently discovered superconducting systems ($T_{\rm c}\sim0.9-2.5$ K) in which the vanadium atoms adopt the kagome structure. Intriguingly, these systems enter a charge-density-wave (CDW) phase ($T_{\rm CDW}\sim80-100$ K), and further evidence shows that the time-reversal symmetry is broken in the CDW phase. Concurrently, the anomalous Hall effect has been observed in KV$_3$Sb$_5$ and CsV$_3$Sb$_5$ inside the novel CDW phase. Here, we report a comprehensive study of a high-quality RbV$_3$Sb$_5$ single crystal with magnetotransport measurements. Our data demonstrate the emergence of anomalous Hall effect in RbV$_3$Sb$_5$ when the charge-density-wave state develops. The magnitude of anomalous Hall resistivity at the low temperature limit is comparable to the reported values in KV$_3$Sb$_5$ and CsV$_3$Sb$_5$. The magnetoresistance channel further reveals a rich spectrum of quantum oscillation frequencies, many of which have not been reported before. In particular, a large quantum oscillation frequency (2235 T), which occupies $\sim$56% of the Brillouin zone area, has been recorded. For the quantum oscillation frequencies with sufficient signal-to-noise ratio, we further perform field-angle dependent measurements and our data indicate two-dimensional Fermi surfaces in RbV$_3$Sb$_5$. Our results provide indispensable information for understanding the anomalous Hall effect and band structure in kagome metals AV$_3$Sb$_5$., Comment: 8 pages, 5 figures. J. Phys. Mater. (in press)

Quantum computers have the potential to efficiently simulate large-scale quantum systems for which classical approaches are bound to fail. Even though several existing quantum devices now feature total qubit numbers of more than one hundred, their applicability remains plagued by the presence of noise and errors. Thus, the degree to which large quantum systems can successfully be simulated on these devices remains unclear. Here, we report on cloud simulations performed on several of IBM's superconducting quantum computers to simulate ground states of spin chains having a wide range of system sizes up to one hundred and two qubits. We find that the ground-state energies extracted from realizations across different quantum computers and system sizes reach the expected values to within errors that are small (i.e. on the percent level), including the inference of the energy density in the thermodynamic limit from these values. We achieve this accuracy through a combination of physics-motivated variational Ansatzes, and efficient, scalable energy-measurement and error-mitigation protocols, including the use of a reference state in the zero-noise extrapolation. By using a 102-qubit system, we have been able to successfully apply up to 3186 CNOT gates in a single circuit when performing gate-error mitigation. Our accurate, error-mitigated results for random parameters in the Ansatz states suggest that a standalone hybrid quantum-classical variational approach for large-scale XXZ models is feasible., Comment: 21 pages, 12 figures, 4 tables; title change; substantial revision

Vision-language navigation is the task of directing an embodied agent to navigate in 3D scenes with natural language instructions. For the agent, inferring the long-term navigation target from visual-linguistic clues is crucial for reliable path planning, which, however, has rarely been studied before in literature. In this article, we propose a Target-Driven Structured Transformer Planner (TD-STP) for long-horizon goal-guided and room layout-aware navigation. Specifically, we devise an Imaginary Scene Tokenization mechanism for explicit estimation of the long-term target (even located in unexplored environments). In addition, we design a Structured Transformer Planner which elegantly incorporates the explored room layout into a neural attention architecture for structured and global planning. Experimental results demonstrate that our TD-STP substantially improves previous best methods' success rate by 2% and 5% on the test set of R2R and REVERIE benchmarks, respectively. Our code is available at https://github.com/YushengZhao/TD-STP .

The emergence of high transition temperature ($T_c$) superconductivity in bulk FeSe under pressure is associated with the tuning of nematicity and magnetism. However, sorting out the relative contributions from magnetic and nematic fluctuations to the enhancement of $T_c$ remains challenging. Here, we design and conduct a series of high-pressure experiments on FeSe thin flakes. We find that, as the thickness decreases, the nematic phase boundary on temperature-pressure phase diagrams remains robust while the magnetic order is significantly weakened. A local maximum of $T_c$ is observed outside the nematic phase region, not far from the extrapolated nematic endpoint in all samples. However, the maximum $T_c$ value is reduced associated with the weakening of magnetism. No high-$T_c$ phase is observed in the thinnest sample. Our results strongly suggest that nematic fluctuations alone can only have a limited effect while magnetic fluctuations are pivotal on the enhancement of $T_c$ in FeSe., Comment: 8 pages, 5 figures. This is the first draft of an accepted paper by Nano Letters

Recently proposed fine-grained 3D visual grounding is an essential and challenging task, whose goal is to identify the 3D object referred by a natural language sentence from other distractive objects of the same category. Existing works usually adopt dynamic graph networks to indirectly model the intra/inter-modal interactions, making the model difficult to distinguish the referred object from distractors due to the monolithic representations of visual and linguistic contents. In this work, we exploit Transformer for its natural suitability on permutation-invariant 3D point clouds data and propose a TransRefer3D network to extract entity-and-relation aware multimodal context among objects for more discriminative feature learning. Concretely, we devise an Entity-aware Attention (EA) module and a Relation-aware Attention (RA) module to conduct fine-grained cross-modal feature matching. Facilitated by co-attention operation, our EA module matches visual entity features with linguistic entity features while RA module matches pair-wise visual relation features with linguistic relation features, respectively. We further integrate EA and RA modules into an Entity-and-Relation aware Contextual Block (ERCB) and stack several ERCBs to form our TransRefer3D for hierarchical multimodal context modeling. Extensive experiments on both Nr3D and Sr3D datasets demonstrate that our proposed model significantly outperforms existing approaches by up to 10.6% and claims the new state-of-the-art. To the best of our knowledge, this is the first work investigating Transformer architecture for fine-grained 3D visual grounding task., Comment: ACM MM2021

Pressure calibration for most diamond-anvil cell (DAC) experiments is mainly based on the ruby scale, which is key to implement this powerful tool for high-pressure study. However, the ruby scale can often hardly be used for programmably-controlled DAC devices, especially the piezoelectric-driving cells, where a continuous pressure calibration is required. In this work, we present an effective pressure gauge for DACs made of manganin metal, based on the four-probe resistivity measurements. Pressure dependence of its resistivity is well established and shows excellent linear relations in the 0 - 30 GPa pressure range with a slope of 23.4 (9) GPa for the first-cycle compression, in contrast to that of multiple-cycle compression and decompression having a nearly identical slope of 33.7 (4) GPa likely due to the strain effect. In addition, such-established manganin scale can be used for continuously monitoring the cell pressure of piezoelectric-driving DACs, and the reliability of this method is also verified by the fixed-point method with a Bi pressure standard. Realization of continuous pressure calibration for programmably-controlled DACs would offer many opportunities for study of dynamics, kinetics, and critical behaviors of pressure-induced phase transitions.

Coupling of charge and lattice degrees of freedom in materials can produce intriguing electronic phenomena, such as conventional superconductivity where the electrons are mediated by lattice for creating supercurrent. The Mott transition, which is a source for many fascinating emergent behaviors, is originally thought to be driven solely by correlated electrons with an Ising criticality. Recent studies on the known Mott systems have shown that the lattice degree of freedom is also at play, giving rise to either Landau or unconventional criticality. However, the underlying coupling mechanism of charge and lattice degrees of freedom around the Mott critical endpoint remains elusive, leading to difficulties in understanding the associated Mott physics. Here we report a study of Mott transition in cubic PbCrO3 by measuring the lattice parameter, using high-pressure x-ray diffraction techniques. The Mott criticality in this material is revealed with large lattice anomalies, which is governed by giant viscoelasticity that presumably results from a combination of lattice elasticity and electron viscosity. Because of the viscoelastic effect, the lattice of this material behaves peculiarly near the critical endpoint, inconsistent with any existing university classes. We argue that the viscoelasticity may play as a hidden degree of freedom behind the Mott criticality., Comment: 12 pages main text including 3 figures. 20 pages suppporting material

Deep neural networks have been widely used in many computer vision tasks. However, it is proved that they are susceptible to small, imperceptible perturbations added to the input. Inputs with elaborately designed perturbations that can fool deep learning models are called adversarial examples, and they have drawn great concerns about the safety of deep neural networks. Object detection algorithms are designed to locate and classify objects in images or videos and they are the core of many computer vision tasks, which have great research value and wide applications. In this paper, we focus on adversarial attack on some state-of-the-art object detection models. As a practical alternative, we use adversarial patches for the attack. Two adversarial patch generation algorithms have been proposed: the heatmap-based algorithm and the consensus-based algorithm. The experiment results have shown that the proposed methods are highly effective, transferable and generic. Additionally, we have applied the proposed methods to competition "Adversarial Challenge on Object Detection" that is organized by Alibaba on the Tianchi platform and won top 7 in 1701 teams. Code is available at: https://github.com/FenHua/DetDak

We investigated the evolution of ferromagnetism in layered Fe$_3$GeTe$_2$ flakes under different pressures and temperatures using in situ magnetic circular dichroism (MCD) spectroscopy. We found that the rectangle shape of hysteretic loop under an out-of-plane magnetic field sweep can sustain below 7 GPa. Above that pressure, an intermediate state appears at low temperature region signaled by an 8-shaped skew hysteretic loop. Meanwhile, the coercive field and Curie temperature decrease with increasing pressures, implying the decrease of the exchange interaction and the magneto-crystalline anisotropy under pressures. The intermediate phase has a labyrinthine domain structure, which is attributed to the increase of ratio of exchange interaction to magneto-crystalline anisotropy based on Jagla's theory. Moreover, our calculation results reveal a weak structural transition around 6 GPa, which leads to a drop of the magnetic momentum of Fe ions.

Layered transition-metal compounds with controllable magnetic behaviors provide many fascinating opportunities for the fabrication of high-performance magneto-electric and spintronic devices. The tuning of their electronic and magnetic properties is usually limited to the change of layer thickness, electrostatic doping, and the control of electric and magnetic fields. However, pressure has been rarely exploited as a control parameter for tailoring their magneto-electric properties. Here, we report a unique pressure-driven isostructural phase transition in layered CrCl3 accompanied by a simultaneous switching of magnetism from a ferromagnetic to an antiferromagnetic ordering. Our experiments, in combination with ab initio calculations, demonstrate that such a magnetic transformation hinders the band-gap collapse under pressure, leading to an anomalous semiconductor-to-semiconductor transition. Our findings not only reveal high potential of CrCl3 in electronic and spintronic applications under ambient and extreme conditions but also establish the basis for exploring unusual phase transitions in layered transition-metal compounds.

We perform core-level X-ray photoemission spectroscopy (XPS) and electronic Raman scattering studies of electronic structures and spin fluctuations in the bulk samples of the nickelate oxide NdNiO$_2$. According to Nd $3d$ and O $1s$ XPS spectra, we conclude that NdNiO$_2$ has a large transfer energy. From the analysis of the main line of the Ni $2p_{3/2}$ XPS, we confirm the NiO$_2$ planes in NdNiO$_2$ are of Mott-Hubbard type in the Zaanen-Sawatzky-Allen scheme. The two-magnon peak in the Raman scattering provides direct evidence for the strong spin-fluctuation in NdNiO$_2$. The peak position determines the antiferromagnetic exchange $J=25$~meV. Our experimental results agree well with our previous theoretical results., Comment: 7 pages, 5 figures

Structural symmetry of crystals plays important roles in physical properties of two-dimensional (2D) materials, particularly in the nonlinear optics regime. It has been a long-term exploration on the physical properties in 2D materials with various stacking structures, which correspond to different structural symmetries. Usually, the manipulation of rotational alignment between layers in 2D heterostructures has been realized at the synthetic stage through artificial stacking like assembling Lego bricks. However, the reconfigurable control of translational symmetry of crystalline structure is still challenging. High pressure, as a powerful external control knob, provides a very promising route to circumvent this constraint. Here, we experimentally demonstrate a pressure-controlled symmetry transition in layered InSe. The continuous and reversible evolution of structural symmetries can be in-situ monitored by using the polarization-resolved second harmonic generation (SHG) spectroscopy. As pressure changes, the reconfigurable symmetry transition of the SHG pattern from three-fold rotational symmetry to mirror symmetry was experimentally observed in a layered InSe samples and was successfully explained by the proposed interlayer-translation model. This opens new routes towards potential applications of manipulating crystal symmetry of 2D materials.