We consider a new kind of clustering problem in which clusters need not be independent of each other, but rather can have compositional relationships with other clusters (e.g., an image set consists of rectangles, circles, as well as combinations of
rectangles and circles). This task is motivated by recent work in few-shot learning on compositional embedding models that structure the embedding space to distinguish the label sets, not just the individual labels, assigned to the examples. To tackle this clustering problem, we propose a new algorithm called Compositional Affinity Propagation (CAP). In contrast to standard Affinity Propagation as well as other algorithms for multi-view and hierarchical clustering, CAP can deduce compositionality among clusters automatically. We show promising results, compared to several existing clustering algorithms, on the MultiMNIST, OmniGlot, and LibriSpeech datasets. Our work has applications to multi-object image recognition and speaker diarization with simultaneous speech from multiple speakers.
We prove the long-standing Payne conjecture that the $k^{text{th}}$ eigenvalue in the buckling problem for a clamped plate is not less than the ${k+1}^{text{st}}$ eigenvalue for the membrane of the same shape which is fixed on the boundary. Moreover,
we show that the Payne conjecture is still true for $n$-dimensional case ($nge 2)$.
The LAMOST Medium-Resolution Spectroscopic Survey (LAMOST-MRS) provides an unprecedented opportunity for detecting multi-line spectroscopic systems. Based on the method of Cross-Correlation Function (CCF) and successive derivatives, we search for spe
ctroscopic binaries and triples and derive their radial velocities (RVs) from the LAMOST-MRS spectra. A Monte-Carlo simulation is adopted to estimate the RV uncertainties. After examining over 1.3 million LAMOST DR7 MRS blue arm spectra, we obtain 3,133 spectroscopic binary (SB) and 132 spectroscopic triple (ST) candidates, which account for 1.2% of the LAMOST-MRS stars. Over 95% of the candidates are newly discovered. It is found that all of the ST candidates are on the main sequence, while around 10% of the SB candidates may have one or two components on the red giant branch.
Event extraction (EE), which acquires structural event knowledge from texts, can be divided into two sub-tasks: event type classification and element extraction (namely identifying triggers and arguments under different role patterns). As different e
vent types always own distinct extraction schemas (i.e., role patterns), previous work on EE usually follows an isolated learning paradigm, performing element extraction independently for different event types. It ignores meaningful associations among event types and argument roles, leading to relatively poor performance for less frequent types/roles. This paper proposes a novel neural association framework for the EE task. Given a document, it first performs type classification via constructing a document-level graph to associate sentence nodes of different types, and adopting a graph attention network to learn sentence embeddings. Then, element extraction is achieved by building a universal schema of argument roles, with a parameter inheritance mechanism to enhance role preference for extracted elements. As such, our model takes into account type and role associations during EE, enabling implicit information sharing among them. Experimental results show that our approach consistently outperforms most state-of-the-art EE methods in both sub-tasks. Particularly, for types/roles with less training data, the performance is superior to the existing methods.
This paper is devoted to investigate the heat trace asymptotic expansion corresponding to the magnetic Steklov eigenvalue problem on Riemannian manifolds with boundary. We establish an effective procedure, by which we can calculate all the coefficien
ts $a_0$, $a_1$, $dots$, $a_{n-1}$ of the heat trace asymptotic expansion. In particular, we explicitly give the expressions for the first four coefficients. These coefficients are spectral invariants which provide precise information concerning the volume and curvatures of the boundary of the manifold and some physical quantities by the magnetic Steklov eigenvalues.
Dual-functional radar-communication (DFRC) systems can simultaneously perform both radar and communication functionalities using the same hardware platform and spectrum resource. In this paper, we consider multi-input multi-output (MIMO) DFRC systems
and focus on transmit beamforming designs to provide both radar sensing and multi-user communications. Unlike conventional block-level precoding techniques, we propose to use the recently emerged symbol-level precoding approach in DFRC systems, which provides additional degrees of freedom (DoFs) that guarantee preferable instantaneous transmit beampatterns for radar sensing and achieve better communication performance. In particular, the squared error between the designed and desired beampatterns is minimized subject to the quality-of-service (QoS) requirements of the communication users and the constant-modulus power constraint. Two efficient algorithms are developed to solve this non-convex problem on both the Euclidean and Riemannian spaces. The first algorithm employs penalty dual decomposition (PDD), majorization-minimization (MM), and block coordinate descent (BCD) methods to convert the original optimization problem into two solvable sub-problems, and iteratively solves them using efficient algorithms. The second algorithm provides a much faster solution at the price of a slight performance loss, first transforming the original problem into Riemannian space, and then utilizing the augmented Lagrangian method (ALM) to obtain an unconstrained problem that is subsequently solved via a Riemannian Broyden-Fletcher-Goldfarb-Shanno (RBFGS) algorithm. Extensive simulations verify the distinct advantages of the proposed symbol-level precoding designs in both radar sensing and multi-user communications.
Improving lithium-ion batteries (LIBs) safety remains in a challenging task when compared with the tremendous progress made in their performance in recent years. Embedding thermo-responsive polymer switching materials (TRPS) into LIB cells has been p
roved to be a promising strategy to provide consistent thermal abuse protections at coin-cell level. However, it is unrealistic to achieve large-scale applications without further demonstration in high-capacity pouch cells. Here, we employed tungsten carbide (WC) as a novel conductive filler, and successfully overcame the intrinsic processing difficulty of polyethylene (PE) matrix in a scalable solvent-based method to obtain ultra-thin, uniform, highly conductive TRPS. Moreover, by integrating TRPS directly into LIB electrodes, no extra fabrication facilities or processes are required for making the cells. As a result, multi-layer pouch cells with consistent electrochemical performance and thermal abuse protection function were fabricated using industry relevant manufacturing techniques, which brings TRPS one step further to the real application scenarios.
Language instruction plays an essential role in the natural language grounded navigation tasks. However, navigators trained with limited human-annotated instructions may have difficulties in accurately capturing key information from the complicated i
nstruction at different timesteps, leading to poor navigation performance. In this paper, we exploit to train a more robust navigator which is capable of dynamically extracting crucial factors from the long instruction, by using an adversarial attacking paradigm. Specifically, we propose a Dynamic Reinforced Instruction Attacker (DR-Attacker), which learns to mislead the navigator to move to the wrong target by destroying the most instructive information in instructions at different timesteps. By formulating the perturbation generation as a Markov Decision Process, DR-Attacker is optimized by the reinforcement learning algorithm to generate perturbed instructions sequentially during the navigation, according to a learnable attack score. Then, the perturbed instructions, which serve as hard samples, are used for improving the robustness of the navigator with an effective adversarial training strategy and an auxiliary self-supervised reasoning task. Experimental results on both Vision-and-Language Navigation (VLN) and Navigation from Dialog History (NDH) tasks show the superiority of our proposed method over state-of-the-art methods. Moreover, the visualization analysis shows the effectiveness of the proposed DR-Attacker, which can successfully attack crucial information in the instructions at different timesteps. Code is available at https://github.com/expectorlin/DR-Attacker.
Recent years pre-trained language models hit a success on modeling natural language sentences and (semi-)structured tables. However, existing table pre-training techniques always suffer from low data quality and low pre-training efficiency. In this p
aper, we show that table pre-training can be realized by learning a neural SQL executor over a synthetic corpus, which is obtained by automatically synthesizing executable SQL queries. By pre-training on the synthetic corpus, our approach TAPEX dramatically improves the performance on downstream tasks, boosting existing language models by at most 19.5%. Meanwhile, TAPEX has remarkably high pre-training efficiency and yields strong results when using a small pre-trained corpus. Experimental results demonstrate that TAPEX outperforms previous table pre-training approaches by a large margin, and our model achieves new state-of-the-art results on four well-known datasets, including improving the WikiSQL denotation accuracy to 89.6% (+4.9%), the WikiTableQuestions denotation accuracy to 57.5% (+4.8%), the SQA denotation accuracy to 74.5% (+3.5%), and the TabFact accuracy to 84.6% (+3.6%). Our work opens the way to reason over structured data by pre-training on synthetic executable programs.
In the paper, we study the plane Couette flow of a rarefied gas between two parallel infinite plates at $y=pm L$ moving relative to each other with opposite velocities $(pm alpha L,0,0)$ along the $x$-direction. Assuming that the stationary state tak
es the specific form of $F(y,v_x-alpha y,v_y,v_z)$ with the $x$-component of the molecular velocity sheared linearly along the $y$-direction, such steady flow is governed by a boundary value problem on a steady nonlinear Boltzmann equation driven by an external shear force under the homogeneous non-moving diffuse reflection boundary condition. In case of the Maxwell molecule collisions, we establish the existence of spatially inhomogeneous non-equilibrium stationary solutions to the steady problem for any small enough shear rate $alpha>0$ via an elaborate perturbation approach using Caflischs decomposition together with Guos $L^inftycap L^2$ theory. The result indicates the polynomial tail at large velocities for the stationary distribution. Moreover, the large time asymptotic stability of the stationary solution with an exponential convergence is also obtained and as a consequence the nonnegativity of the steady profile is justified.
