Search

Zero-shot graph machine learning, especially with graph neural networks (GNNs), has garnered significant interest due to the challenge of scarce labeled data. While methods like self-supervised learning and graph prompt learning have been extensively explored, they often rely on fine-tuning with task-specific labels, limiting their effectiveness in zero-shot scenarios. Inspired by the zero-shot capabilities of instruction-fine-tuned large language models (LLMs), we introduce a novel framework named Token Embedding-Aligned Graph Language Model (TEA-GLM) that leverages LLMs as cross-dataset and cross-task zero-shot learners for graph machine learning. Concretely, we pretrain a GNN, aligning its representations with token embeddings of an LLM. We then train a linear projector that transforms the GNN's representations into a fixed number of graph token embeddings without tuning the LLM. A unified instruction is designed for various graph tasks at different levels, such as node classification (node-level) and link prediction (edge-level). These design choices collectively enhance our method's effectiveness in zero-shot learning, setting it apart from existing methods. Experiments show that our graph token embeddings help the LLM predictor achieve state-of-the-art performance on unseen datasets and tasks compared to other methods using LLMs as predictors.

Commencing from the centrosymmetric MnRMnSbO$_6$ compound, we explore the realm of magnetic polar double-double perovskite oxides characterized by significant ferroelectric polarization. Employing symmetry operations, first-principles methodologies, and Monte Carlo simulations, our investigation delves into the structural, magnetic, ferroelectric, and electronic attributes of the polar LaFeMnNiO$_6$ and LaTiMnNiO$_6$ compounds. The structural analysis uncovers that the paraelectric-ferroelectric phase transition is intricately linked to the Fe/Ti-displacement of square planar Fe/TiO$_4$. Notably, the magnetic LaFeMnNiO$_6$ and LaTiMnNiO$_6$ compounds demonstrate robust ferroelectric polarizations, measuring 20.0 $\mu$C/cm$^2$ and 21.8 $\mu$C/cm$^2$, respectively, accompanied by minimalist forbidden energy gaps of 1.40 eV and 1.18 eV using the GGA+U method. Furthermore, we pinpoint elevated magnetic transition temperatures for these compounds. Additionally, our study scrutinizes the energies associated with diverse spin configurations and identifies potential minimum decomposition pathways into stable oxides. This comprehensive analysis ensures the meticulous formation of the LaFeMnNiO$_6$ and LaTiMnNiO$_6$ compounds.

Accessibility measures how well a location is connected to surrounding opportunities. We focus on accessibility provided by Public Transit (PT). There is an evident inequality in the distribution of accessibility between city centers or close to main transportation corridors and suburbs. In the latter, poor PT service leads to a chronic car-dependency. Demand-Responsive Transit (DRT) is better suited for low-density areas than conventional fixed-route PT. However, its potential to tackle accessibility inequality has not yet been exploited. On the contrary, planning DRT without care to inequality (as in the methods proposed so far) can further improve the accessibility gap in urban areas. To the best of our knowledge this paper is the first to propose a DRT planning strategy, which we call AccEq-DRT, aimed at reducing accessibility inequality, while ensuring overall efficiency. To this aim, we combine a graph representation of conventional PT and a Continuous Approximation (CA) model of DRT. The two are combined in the same multi-layer graph, on which we compute accessibility. We then devise a scoring function to estimate the need of each area for an improvement, appropriately weighting population density and accessibility. Finally, we provide a bilevel optimization method, where the upper level is a heuristic to allocate DRT buses, guided by the scoring function, and the lower level performs traffic assignment. Numerical results in a simplified model of Montreal show that inequality, measured with the Atkinson index, is reduced by up to 34\%. Keywords: DRT Public, Transportation, Accessibility, Continuous Approximation, Network Design, Comment: 15 pages

Heterogeneous graph neural networks (HGNNs) have emerged as powerful algorithms for processing heterogeneous graphs (HetGs), widely used in many critical fields. To capture both structural and semantic information in HetGs, HGNNs first aggregate the neighboring feature vectors for each vertex in each semantic graph and then fuse the aggregated results across all semantic graphs for each vertex. Unfortunately, existing graph neural network accelerators are ill-suited to accelerate HGNNs. This is because they fail to efficiently tackle the specific execution patterns and exploit the high-degree parallelism as well as data reusability inside and across the processing of semantic graphs in HGNNs. In this work, we first quantitatively characterize a set of representative HGNN models on GPU to disclose the execution bound of each stage, inter-semantic-graph parallelism, and inter-semantic-graph data reusability in HGNNs. Guided by our findings, we propose a high-performance HGNN accelerator, HiHGNN, to alleviate the execution bound and exploit the newfound parallelism and data reusability in HGNNs. Specifically, we first propose a bound-aware stage-fusion methodology that tailors to HGNN acceleration, to fuse and pipeline the execution stages being aware of their execution bounds. Second, we design an independency-aware parallel execution design to exploit the inter-semantic-graph parallelism. Finally, we present a similarity-aware execution scheduling to exploit the inter-semantic-graph data reusability. Compared to the state-of-the-art software framework running on NVIDIA GPU T4 and GPU A100, HiHGNN respectively achieves an average 41.5$\times$ and 8.6$\times$ speedup as well as 106$\times$ and 73$\times$ energy efficiency with quarter the memory bandwidth of GPU A100., Comment: 16 pages, 17 figures; To appear in IEEE TPDS 2024

Speech generation for machine dubbing adds complexity to conventional Text-To-Speech solutions as the generated output is required to match the expressiveness, emotion and speaking rate of the source content. Capturing and transferring details and variations in prosody is a challenge. We introduce phrase-level cross-lingual prosody transfer for expressive multi-lingual machine dubbing. The proposed phrase-level prosody transfer delivers a significant 6.2% MUSHRA score increase over a baseline with utterance-level global prosody transfer, thereby closing the gap between the baseline and expressive human dubbing by 23.2%, while preserving intelligibility of the synthesised speech., Comment: Accepted to INTERSPEECH 2023

Prosody transfer is well-studied in the context of expressive speech synthesis. Cross-lingual prosody transfer, however, is challenging and has been under-explored to date. In this paper, we present a novel solution to learn prosody representations that are transferable across languages and speakers for machine dubbing of expressive multimedia contents. Multimedia contents often contain field recordings. To enable prosody transfer from noisy audios, we introduce a novel noise modelling module that disentangles noise conditioning from prosody conditioning, and thereby gains independent control of noise levels in the synthesised speech. We augment noisy training data with clean data to improve the ability of the model to map the denoised reference audio to clean speech. Our proposed system can generate speech with context-matching prosody and closes the gap between a strong baseline and human expressive dialogs by 11.2%., Comment: Submitted to INTERSPEECH

The orbital, which represents the shape of the electron cloud, very often strongly influences the manifestation of various exotic phenomena, e.g., magnetism, metal-insulator transition, colossal magnetoresistance, unconventional superconductivity etc. in solid-state systems. The observation of the antiferromagnetism in $RE$TiO$_3$ ($RE$=rare earth) series has been puzzling since the celebrated Kugel-Khomskii model of spin-orbital super exchange predicts ferromagnetism in an orbitally degenerate $d^1$ systems. Further, the existence of the orbitally ordered vs. orbital liquid phase in both antiferromagnetic and paramagnetic phase have been unsettled issues thus far. To address these long-standing questions, we investigate single crystalline film of PrTiO$_3$. Our synchrotron X-ray diffraction measurements confirm the retention of bulk-like orthorhombic ($D_{2h}$) symmetry in the thin film geometry. We observe similar X-ray linear dichroism signal in both paramagnetic and antiferromagnetic phase, which can be accounted by ferro orbital ordering (FOO). While the presence of $D_{2h}$ crystal field does not guarantee lifting of orbital degeneracy always, we find it to be strong enough in these rare-earth titanates, leading to the FOO state. Thus, our work demonstrates the orthorhombic distortion is the driving force for the orbital ordering of antiferromagnetic $RE$TiO$_3$., Comment: 37 pages and 17 figures

Limited by the memory capacity and compute power, singe-node graph convolutional neural network (GCN) accelerators cannot complete the execution of GCNs within a reasonable amount of time, due to the explosive size of graphs nowadays. Thus, large-scale GCNs call for a multi-node acceleration system (MultiAccSys) like TPU-Pod for large-scale neural networks. In this work, we aim to scale up single-node GCN accelerators to accelerate GCNs on large-scale graphs. We first identify the communication pattern and challenges of multi-node acceleration for GCNs on large-scale graphs. We observe that (1) coarse-grained communication patterns exist in the execution of GCNs in MultiAccSys, which introduces massive amount of redundant network transmissions and off-chip memory accesses; (2) overall, the acceleration of GCNs in MultiAccSys is bandwidth-bound and latency-tolerant. Guided by these two observations, we then propose MultiGCN, the first MultiAccSys for large-scale GCNs that trades network latency for network bandwidth. Specifically, by leveraging the network latency tolerance, we first propose a topology-aware multicast mechanism with a one put per multicast message-passing model to reduce transmissions and alleviate network bandwidth requirements. Second, we introduce a scatter-based round execution mechanism which cooperates with the multicast mechanism and reduces redundant off-chip memory accesses. Compared to the baseline MultiAccSys, MultiGCN achieves 4~12x speedup using only 28%~68% energy, while reducing 32% transmissions and 73% off-chip memory accesses on average. It not only achieves 2.5~8x speedup over the state-of-the-art multi-GPU solution, but also scales to large-scale graphs as opposed to single-node GCN accelerators., Comment: To appear in TC

Previous graph analytics accelerators have achieved great improvement on throughput by alleviating irregular off-chip memory accesses. However, on-chip side datapath conflicts and design centralization have become the critical issues hindering further throughput improvement. In this paper, a general solution, Multiple-stage Decentralized Propagation network (MDP-network), is proposed to address these issues, inspired by the key idea of trading latency for throughput. Besides, a novel High throughput Graph analytics accelerator, HiGraph, is proposed by deploying MDP-network to address each issue in practice. The experiment shows that compared with state-of-the-art accelerator, HiGraph achieves up to 2.2x speedup (1.5x on average) as well as better scalability., Comment: To Appear in 59th Design Automation Conference (DAC 2022)

Bayesian Neural Networks (BNNs) offer a mathematically grounded framework to quantify the uncertainty of model predictions but come with a prohibitive computation cost for both training and inference. In this work, we show a novel network architecture search (NAS) that optimizes BNNs for both accuracy and uncertainty while having a reduced inference latency. Different from canonical NAS that optimizes solely for in-distribution likelihood, the proposed scheme searches for the uncertainty performance using both in- and out-of-distribution data. Our method is able to search for the correct placement of Bayesian layer(s) in a network. In our experiments, the searched models show comparable uncertainty quantification ability and accuracy compared to the state-of-the-art (deep ensemble). In addition, the searched models use only a fraction of the runtime compared to many popular BNN baselines, reducing the inference runtime cost by $2.98 \times$ and $2.92 \times$ respectively on the CIFAR10 dataset when compared to MCDropout and deep ensemble.

Multi-Modal Self-Supervised Learning from videos has been shown to improve model's performance on various downstream tasks. However, such Self-Supervised pre-training requires large batch sizes and a large amount of computation resources due to the noise present in the uncurated data. This is partly due to the fact that the prevalent training scheme is trained on coarse-grained setting, in which vectors representing the whole video clips or natural language sentences are used for computing similarity. Such scheme makes training noisy as part of the video clips can be totally not correlated with the other-modality input such as text description. In this paper, we propose a fine-grained multi-modal self-supervised training scheme that computes the similarity between embeddings at finer-scale (such as individual feature map embeddings and embeddings of phrases), and uses attention mechanisms to reduce noisy pairs' weighting in the loss function. We show that with the proposed pre-training scheme, we can train smaller models, with smaller batch-size and much less computational resources to achieve downstream tasks performances comparable to State-Of-The-Art, for tasks including action recognition and text-image retrievals., Comment: Accepted at BMVC 2021

We provide strong numerical evidence for a hidden multihyperuniform long-range order (MHLRO) in SiGeSn medium-entropy alloys (MEAs), in which the normalized infinite-wavelength composition fluctuations for all three atomic species are completely suppressed as in a perfect crystalline state. We show this MHLRO naturally leads to the emergence of short-range order (SRO) recently discovered in MEAs, which results in stable lower-energy states compared to alloy models with random or special quasi-random structures (SQSs) possessing no atomic SROs. The MHLRO MEAs approximately realize the Vegard's law, which offers a rule-of-mixture type predictions of the lattice constants and electronic band gap, and thus can be considered as an ideal mixing state. The MHLRO also directly gives rise to enhanced electronic band gaps and superior thermal transport properties at low temperatures compared to random structures and SQSs, which open up novel potential applications in optoelectronics and thermoelectrics. Our analysis of the SiGeSn system leads to the formulation of general organizing principles applicable in other medium- and high-entropy alloys (HEAs), and a highly efficient computational model for rendering realistic large-scale configurations of MEAs and HEAs.

Atomically thin two dimensional magnets have given rise to emergent phenomena due to magnetic exchange and spin-orbit coupling showing a great promise for realizing ultrathin device structures. In this paper, we critically examine the magnetic properties of 2D FeS$_2$, which has been recently claimed to exhibit room temperature ferromagnetism in the (111) orientation. Our ab initio study based on collinear density functional theory has revealed the ground state as an antiferromagnetic one with an ordering temperature of around 100K along with a signature of spin-phonon coupling, which may trigger a ferromagnetic coupling via strain. Moreover, our calculations based on spin-spirals indicate the possibility of non-collinear magnetic structures, which is also supported by Monte Carlo simulations based on ab initio magnetic exchange parameters. This opens up an excellent possibility to manipulate magnetic structures by the application of directional strain., Comment: 6 pages, 3 figures

The movement of subaqueous sediment in laminar shearing flow is numerically investigated by the coupled lattice Boltzmann and discrete element methods. First, the numerical method is validated by comparing the phase diagram proposed by Ouriemi et al. ({\it J. Fluid Mech}., vol. 636, 2009, pp. 321-336). Second, a detailed study on sediment movement is performed for sediment with varying solid volume fractions, and a nonlinear relationship between the normalised thickness of the mobile layer and the normalised fluid flow rate is observed for a densely-packed sediment. Third, an independent investigation on the effective viscosity and friction coefficient of the sediment under different fluid flow rates is conducted in a shear cell; and substitution of these two critical parameters into a theoretical expression proposed by Aussillous et al. ({\it J. Fluid Mech}., vol. 736, 2013, pp. 594-615) provides consistent predictions of bedload thickness with the simulation results of sediment movement. Therefore, we conclude that the non-Newtonian behaviour of densely-packed sediment leads to the nonlinear relationship between the normalised thickness of the mobile layer and the normalised fluid flow rate.