arXiv subjects

Tianle Wang

Publications and source records attributed to Tianle Wang.

50 records · Page 3Linked to original sources

arXiv2024

MusicAOG: an Energy-Based Model for Learning and Sampling a Hierarchical Representation of Symbolic Music

In addressing the challenge of interpretability and generalizability of artificial music intelligence, this paper introduces a novel symbolic representation that amalgamates both explicit and implicit musical information across diverse traditions and granularities. Utilizing a hierarchical and-or graph representation, the model employs nodes and edges to encapsulate a broad spectrum of musical elements, including structures, textures, rhythms, and harmonies. This hierarchical approach expands the representability across various scales of music. This representation serves as the foundation for an energy-based model, uniquely tailored to learn musical concepts through a flexible algorithm framework relying on the minimax entropy principle. Utilizing an adapted Metropolis-Hastings sampling technique, the model enables fine-grained control over music generation. A comprehensive empirical evaluation, contrasting this novel approach with existing methodologies, manifests considerable advancements in interpretability and controllability. This study marks a substantial contribution to the fields of music analysis, composition, and computational musicology.

cs.SD↗

arXiv2023

WOT-Class: Weakly Supervised Open-world Text Classification

State-of-the-art weakly supervised text classification methods, while significantly reduced the required human supervision, still requires the supervision to cover all the classes of interest. This is never easy to meet in practice when human explore new, large corpora without complete pictures. In this paper, we work on a novel yet important problem of weakly supervised open-world text classification, where supervision is only needed for a few examples from a few known classes and the machine should handle both known and unknown classes in test time. General open-world classification has been studied mostly using image classification; however, existing methods typically assume the availability of sufficient known-class supervision and strong unknown-class prior knowledge (e.g., the number and/or data distribution). We propose a novel framework WOT-Class that lifts those strong assumptions. Specifically, it follows an iterative process of (a) clustering text to new classes, (b) mining and ranking indicative words for each class, and (c) merging redundant classes by using the overlapped indicative words as a bridge. Extensive experiments on 7 popular text classification datasets demonstrate that WOT-Class outperforms strong baselines consistently with a large margin, attaining 23.33% greater average absolute macro-F1 over existing approaches across all datasets. Such competent accuracy illuminates the practical potential of further reducing human effort for text classification.

cs.CL↗

arXiv2023

Denoised Self-Augmented Learning for Social Recommendation

Social recommendation is gaining increasing attention in various online applications, including e-commerce and online streaming, where social information is leveraged to improve user-item interaction modeling. Recently, Self-Supervised Learning (SSL) has proven to be remarkably effective in addressing data sparsity through augmented learning tasks. Inspired by this, researchers have attempted to incorporate SSL into social recommendation by supplementing the primary supervised task with social-aware self-supervised signals. However, social information can be unavoidably noisy in characterizing user preferences due to the ubiquitous presence of interest-irrelevant social connections, such as colleagues or classmates who do not share many common interests. To address this challenge, we propose a novel social recommender called the Denoised Self-Augmented Learning paradigm (DSL). Our model not only preserves helpful social relations to enhance user-item interaction modeling but also enables personalized cross-view knowledge transfer through adaptive semantic alignment in embedding space. Our experimental results on various recommendation benchmarks confirm the superiority of our DSL over state-of-the-art methods. We release our model implementation at: https://github.com/HKUDS/DSL.

cs.IR↗

arXiv2023

Evaluating Portable Parallelization Strategies for Heterogeneous Architectures in High Energy Physics

High-energy physics (HEP) experiments have developed millions of lines of code over decades that are optimized to run on traditional x86 CPU systems. However, we are seeing a rapidly increasing fraction of floating point computing power in leadership-class computing facilities and traditional data centers coming from new accelerator architectures, such as GPUs. HEP experiments are now faced with the untenable prospect of rewriting millions of lines of x86 CPU code, for the increasingly dominant architectures found in these computational accelerators. This task is made more challenging by the architecture-specific languages and APIs promoted by manufacturers such as NVIDIA, Intel and AMD. Producing multiple, architecture-specific implementations is not a viable scenario, given the available person power and code maintenance issues. The Portable Parallelization Strategies team of the HEP Center for Computational Excellence is investigating the use of Kokkos, SYCL, OpenMP, std::execution::parallel and alpaka as potential portability solutions that promise to execute on multiple architectures from the same source code, using representative use cases from major HEP experiments, including the DUNE experiment of the Long Baseline Neutrino Facility, and the ATLAS and CMS experiments of the Large Hadron Collider. This cross-cutting evaluation of portability solutions using real applications will help inform and guide the HEP community when choosing their software and hardware suites for the next generation of experimental frameworks. We present the outcomes of our studies, including performance metrics, porting challenges, API evaluations, and build system integration.

hep-ex↗

arXiv2023

A Benchmark on Extremely Weakly Supervised Text Classification: Reconcile Seed Matching and Prompting Approaches

Etremely Weakly Supervised Text Classification (XWS-TC) refers to text classification based on minimal high-level human guidance, such as a few label-indicative seed words or classification instructions. There are two mainstream approaches for XWS-TC, however, never being rigorously compared: (1) training classifiers based on pseudo-labels generated by (softly) matching seed words (SEED) and (2) prompting (and calibrating) language models using classification instruction (and raw texts) to decode label words (PROMPT). This paper presents the first XWS-TC benchmark to compare the two approaches on fair grounds, where the datasets, supervisions, and hyperparameter choices are standardized across methods. Our benchmarking results suggest that (1) Both SEED and PROMPT approaches are competitive and there is no clear winner; (2) SEED is empirically more tolerant than PROMPT to human guidance (e.g., seed words, classification instructions, and label words) changes; (3) SEED is empirically more selective than PROMPT to the pre-trained language models; (4) Recent SEED and PROMPT methods have close connections and a clustering post-processing step based on raw in-domain texts is a strong performance booster to both. We hope this benchmark serves as a guideline in selecting XWS-TC methods in different scenarios and stimulate interest in developing guidance- and model-robust XWS-TC methods. We release the repo at https://github.com/ZihanWangKi/x-TC.

cs.CL↗

arXiv2023

On the Blind Spots of Model-Based Evaluation Metrics for Text Generation

In this work, we explore a useful but often neglected methodology for robustness analysis of text generation evaluation metrics: stress tests with synthetic data. Basically, we design and synthesize a wide range of potential errors and check whether they result in a commensurate drop in the metric scores. We examine a range of recently proposed evaluation metrics based on pretrained language models, for the tasks of open-ended generation, translation, and summarization. Our experiments reveal interesting insensitivities, biases, or even loopholes in existing metrics. For example, we find that BERTScore is confused by truncation errors in summarization, and MAUVE (built on top of GPT-2) is insensitive to errors at the beginning or middle of generations. Further, we investigate the reasons behind these blind spots and suggest practical workarounds for a more reliable evaluation of text generation. We have released our code and data at https://github.com/cloudygoose/blindspot_nlg.

cs.CL↗

arXiv2023

Radiative corrections to leptonic decays using infinite-volume reconstruction

Lattice QCD calculations of leptonic decay constants have now reached sub-percent precision so that isospin-breaking corrections, including QED effects, must be included to fully exploit this precision in determining fundamental quantities, in particular the elements of the Cabibbo-Kobayashi-Maskawa (CKM) matrix, from experimental measurements. A number of collaborations have performed, or are performing, such computations. In this paper we develop a new theoretical framework, based on Infinite-Volume Reconstruction (IVR), for the computation of electromagnetic corrections to leptonic decay widths. In this method, the hadronic correlation functions are first processed theoretically in infinite volume, in such a way that the required matrix elements can be determined non-perturbatively from lattice QCD computations with finite-volume uncertainties which are exponentially small in the volume. The cancellation of infrared divergences in this framework is performed fully analytically. We also outline how this IVR treatment can be extended to determine the QED effects in semi-leptonic kaon decays with a similar degree of accuracy.

hep-lat↗

arXiv2023

Portable Programming Model Exploration for LArTPC Simulation in a Heterogeneous Computing Environment: OpenMP vs. SYCL

The evolution of the computing landscape has resulted in the proliferation of diverse hardware architectures, with different flavors of GPUs and other compute accelerators becoming more widely available. To facilitate the efficient use of these architectures in a heterogeneous computing environment, several programming models are available to enable portability and performance across different computing systems, such as Kokkos, SYCL, OpenMP and others. As part of the High Energy Physics Center for Computational Excellence (HEP-CCE) project, we investigate if and how these different programming models may be suitable for experimental HEP workflows through a few representative use cases. One of such use cases is the Liquid Argon Time Projection Chamber (LArTPC) simulation which is essential for LArTPC detector design, validation and data analysis. Following up on our previous investigations of using Kokkos to port LArTPC simulation in the Wire-Cell Toolkit (WCT) to GPUs, we have explored OpenMP and SYCL as potential portable programming models for WCT, with the goal to make diverse computing resources accessible to the LArTPC simulations. In this work, we describe how we utilize relevant features of OpenMP and SYCL for the LArTPC simulation module in WCT. We also show performance benchmark results on multi-core CPUs, NVIDIA and AMD GPUs for both the OpenMP and the SYCL implementations. Comparisons with different compilers will also be given where appropriate.

hep-ex↗

arXiv2022

Kekulé spiral order in magic-angle graphene: a density matrix renormalization group study

When the two layers of a twisted moiré system are subject to different degrees of strain, the effect is amplified by the inverse twist angle, e.g., by a factor of 50 in magic angle twisted bilayer graphene (TBG). Samples of TBG typically have heterostrains of 0.1-0.7%, increasing the bandwidth of the "flat'' bands by as much as tenfold, placing TBG in an intermediate coupling regime. Here we study the phase diagram of TBG in the presence of heterostrain with unbiased, large-scale density matrix renormalization group calculations (bond dimension $χ=24576$), including all spin and valley degrees of freedom. Working at filling $ν= -3$, we find a strain of $0.05\%$ drives a transition from a quantized anomalous Hall insulator into an incommensurate-Kekulé spiral (IKS) phase. This peculiar order, proposed and studied at mean-field level by Kwan et al (PRX 11, 041063), breaks both valley conservation and translation symmetry $\hat{T}$, but preserves a modified translation symmetry $\hat{T}'$ with moiré-incommensurate phase modulation. Even higher strains drive the system to a fully symmetric metal.

cond-mat.str-el↗

arXiv2022

Discovering new physics in rare kaon decays

The decays and mixing of $K$ mesons are remarkably sensitive to the weak interactions of quarks and leptons at high energies. They provide important tests of the standard model at both first and second order in the Fermi constant $G_F$ and offer a window into possible new phenomena at energies as high as 1,000 TeV. These possibilities become even more compelling as the growing capabilities of lattice QCD make high-precision standard model predictions possible. Here we discuss and attempt to forecast some of these capabilities.

hep-lat↗

arXiv2020

Direct CP violation and the $ΔI=1/2$ rule in $K\toππ$ decay from the Standard Model

We present a lattice QCD calculation of the $ΔI=1/2$, $K\toππ$ decay amplitude $A_0$ and $\varepsilon'$, the measure of direct CP-violation in $K\toππ$ decay, improving our 2015 calculation of these quantities. Both calculations were performed with physical kinematics on a $32^3\times 64$ lattice with an inverse lattice spacing of $a^{-1}=1.3784(68)$ GeV. However, the current calculation includes nearly four times the statistics and numerous technical improvements allowing us to more reliably isolate the $ππ$ ground-state and more accurately relate the lattice operators to those defined in the Standard Model. We find ${\rm Re}(A_0)=2.99(0.32)(0.59)\times 10^{-7}$ GeV and ${\rm Im}(A_0)=-6.98(0.62)(1.44)\times 10^{-11}$ GeV, where the errors are statistical and systematic, respectively. The former agrees well with the experimental result ${\rm Re}(A_0)=3.3201(18)\times 10^{-7}$ GeV. These results for $A_0$ can be combined with our earlier lattice calculation of $A_2$ to obtain ${\rm Re}(\varepsilon'/\varepsilon)=21.7(2.6)(6.2)(5.0) \times 10^{-4}$, where the third error represents omitted isospin breaking effects, and Re$(A_0)$/Re$(A_2) = 19.9(2.3)(4.4)$. The first agrees well with the experimental result of ${\rm Re}(\varepsilon'/\varepsilon)=16.6(2.3)\times 10^{-4}$. A comparison of the second with the observed ratio Re$(A_0)/$Re$(A_2) = 22.45(6)$, demonstrates the Standard Model origin of this "$ΔI = 1/2$ rule" enhancement.

hep-lat↗

arXiv2020

Flat band topology of magic angle graphene on a transition metal dichalcogenide

We consider twisted bilayer graphene on a transition metal dichalcogenide substrate, where proximity-induced spin-orbit coupling significantly alters the eight flat bands which occur near the magic angle. The resulting band structure features a pair of extremely flat bands across most of the mini-Brillouin zone. Further details depend sensitively on the symmetries of the heterostructure; we find semiconducting band structures when all two-fold rotations around in-plane axis are broken, and semi-metallic band structures otherwise. We calculate the Chern numbers of the different isolated bands, and identify the parameter regimes and filling factors where valley Chern insulators and topological insulators are realized. Interestingly, we find that for realistic values of the proximity-induced terms, it is possible to realize a topological insulator protected by time-reversal symmetry by doping two holes or two electrons per superlattice unit cell into the system.

cond-mat.str-el↗

arXiv2019

Update on the improved lattice calculation of direct CP-violation in K decays

We discuss the status of the RBC & UKQCD collaboration's lattice determination of $ε'$, the measure of direct CP-violation in kaon decays, focusing in particular on recent developments in statistical techniques for estimating standard errors and goodness-of-fit metrics for large amounts of data that have correlations both in the temporal coordinate and also in molecular dynamics time. A key result is the formulation of a technique for determining the null distribution of a fit using a bootstrap method that is free from the usual assumptions of independence, large-$n$ and/or normal data, that can for instance be applied to compute p-values even for uncorrelated fits.

hep-lat↗

arXiv2019

Semi-Supervised Brain Lesion Segmentation with an Adapted Mean Teacher Model

Automated brain lesion segmentation provides valuable information for the analysis and intervention of patients. In particular, methods based on convolutional neural networks (CNNs) have achieved state-of-the-art segmentation performance. However, CNNs usually require a decent amount of annotated data, which may be costly and time-consuming to obtain. Since unannotated data is generally abundant, it is desirable to use unannotated data to improve the segmentation performance for CNNs when limited annotated data is available. In this work, we propose a semi-supervised learning (SSL) approach to brain lesion segmentation, where unannotated data is incorporated into the training of CNNs. We adapt the mean teacher model, which is originally developed for SSL-based image classification, for brain lesion segmentation. Assuming that the network should produce consistent outputs for similar inputs, a loss of segmentation consistency is designed and integrated into a self-ensembling framework. Specifically, we build a student model and a teacher model, which share the same CNN architecture for segmentation. The student and teacher models are updated alternately. At each step, the student model learns from the teacher model by minimizing the weighted sum of the segmentation loss computed from annotated data and the segmentation consistency loss between the teacher and student models computed from unannotated data. Then, the teacher model is updated by combining the updated student model with the historical information of teacher models using an exponential moving average strategy. For demonstration, the proposed approach was evaluated on ischemic stroke lesion segmentation, where it improves stroke lesion segmentation with the incorporation of unannotated data.

cs.CV↗