Searcharxiv⌕ Search

arXiv subjects

Vincent Cheung

Publications and source records attributed to Vincent Cheung.

14 recordsLinked to original sources
arXiv2026

CN101 - A Digital Thermodynamic Computer for Generative AI

Thermodynamic computing is an emerging hardware paradigm, in which stochastic physical dynamics serve as the direct computational primitive. The recent explosion of generative AI has only sharpened the search for alternative approaches to compute, and, as we show in this work, thermodynamic computing turns out to be well suited to this space. An important class of methods realises a function as the stationary expectation of an ergodic stochastic process: the answer is encoded in the time-averaged statistics of an equilibrating trajectory. To date, this equilibration-style class has been formulated exclusively through Langevin dynamics, restricting its implementations to analogue substrates and the engineering challenges those bring. In this work, we propose a substrate-independent formalisation of the equilibration-style formulation, in which the only object of design is the dynamical generator L* of an arbitrary ergodic process. The formalisation makes three hardware-level properties of the formulation explicit: the precision of a result is a knob set by how long the dynamics are run, sample averages decompose across independent trajectories, and dependent stages of a computation operate concurrently rather than serially, a property we call sequential parallelism. We instantiate the formalisation by fabricating a prototype digital thermodynamic computing chip, named CN101, that implements the formulation through discrete accumulator dynamics on standard CMOS using stochastic computing principles. We characterise CN101's success across conventional generative AI workloads in the form of VAEs and flow matching, applied to both image generation and scientific problems. Together, the formalisation and its digital instantiation show that the equilibration-style formulation is substrate-independent, and that its computational properties can be exploited on standard digital hardware.

cs.ET↗
arXiv2024

How does the teacher rate? Observations from the NeuroPiano dataset

This paper provides a detailed analysis of the NeuroPiano dataset, which comprise 104 audio recordings of student piano performances accompanied with 2255 textual feedback and ratings given by professional pianists. We offer a statistical overview of the dataset, focusing on the standardization of annotations and inter-annotator agreement across 12 evaluative questions concerning performance quality. We also explore the predictive relationship between audio features and teacher ratings via machine learning, as well as annotations provided for text analysis of the responses.

eess.AS↗
arXiv2024

Imagine yourself: Tuning-Free Personalized Image Generation

Diffusion models have demonstrated remarkable efficacy across various image-to-image tasks. In this research, we introduce Imagine yourself, a state-of-the-art model designed for personalized image generation. Unlike conventional tuning-based personalization techniques, Imagine yourself operates as a tuning-free model, enabling all users to leverage a shared framework without individualized adjustments. Moreover, previous work met challenges balancing identity preservation, following complex prompts and preserving good visual quality, resulting in models having strong copy-paste effect of the reference images. Thus, they can hardly generate images following prompts that require significant changes to the reference image, \eg, changing facial expression, head and body poses, and the diversity of the generated images is low. To address these limitations, our proposed method introduces 1) a new synthetic paired data generation mechanism to encourage image diversity, 2) a fully parallel attention architecture with three text encoders and a fully trainable vision encoder to improve the text faithfulness, and 3) a novel coarse-to-fine multi-stage finetuning methodology that gradually pushes the boundary of visual quality. Our study demonstrates that Imagine yourself surpasses the state-of-the-art personalization model, exhibiting superior capabilities in identity preservation, visual quality, and text alignment. This model establishes a robust foundation for various personalization applications. Human evaluation results validate the model's SOTA superiority across all aspects (identity preservation, text faithfulness, and visual appeal) compared to the previous personalization models.

cs.CV↗
arXiv2024

LLaQo: Towards a Query-Based Coach in Expressive Music Performance Assessment

Research in music understanding has extensively explored composition-level attributes such as key, genre, and instrumentation through advanced representations, leading to cross-modal applications using large language models. However, aspects of musical performance such as stylistic expression and technique remain underexplored, along with the potential of using large language models to enhance educational outcomes with customized feedback. To bridge this gap, we introduce LLaQo, a Large Language Query-based music coach that leverages audio language modeling to provide detailed and formative assessments of music performances. We also introduce instruction-tuned query-response datasets that cover a variety of performance dimensions from pitch accuracy to articulation, as well as contextual performance understanding (such as difficulty and performance techniques). Utilizing AudioMAE encoder and Vicuna-7b LLM backend, our model achieved state-of-the-art (SOTA) results in predicting teachers' performance ratings, as well as in identifying piece difficulty and playing techniques. Textual responses from LLaQo was moreover rated significantly higher compared to other baseline models in a user study using audio-text matching. Our proposed model can thus provide informative answers to open-ended questions related to musical performance from audio data.

eess.AS↗
arXiv2024

Direct quarkonium production in DIS from a joint CGC and NRQCD framework

We compute the differential cross section for direct quarkonium production in high-energy electron-nucleus collisions at small $x$. Our computation is performed within the nonrelativistic QCD factorization formalism that separates the calculation into short distance coefficients and long distance matrix elements that depend on the color and spin of the state. We obtain the short distance coefficients of the production of the heavy quark pair within the framework of the Color Glass Condensate effective field theory, which resums coherent multiple interactions of the heavy quark pair with the nucleus to all orders. Our results are expressed as the convolution of perturbatively calculable perturbative functions with multi-point light-like Wilson line correlators. In the correlation limit, we establish the correspondence between our CGC formulation with calculations employing the transverse momentum dependent (TMD) framework. We extend this correspondence by resumming kinematic power corrections within the improved TMD framework, which interpolates between the TMD formalism and $k_\perp$ factorization formalism. We present a detailed numerical analysis, focusing on $J/\psi$ production in the kinematics accessible at the future Electron-Ion Collider, highlighting the importance of genuine higher-order saturation contributions when the electron collides with a large nucleus.

hep-ph↗
arXiv2024

$J/\psi$ photoproduction and polarization in $e+p$ collisions in the improved color evaporation model

We calculate the production and polarization of direct $J/\psi$ in the improved color evaporation model in $e+p$ photoproduction. We present the production as functions of transverse momentun, mass of the hadronic final state, and inelasticity. We also present the polarization parameters $\lambda_\vartheta$, $\lambda_{\varphi}$, and $\lambda_{\vartheta \varphi}$ in the helicity and the Collins-Soper frames, as well as the frame-invariant polarization parameter $\tilde{\lambda}$ as a function of transverse momentum and inelasticity. We find agreement with both $J/\psi$ unpolarized cross sections and the invariant polarization parameters as a function of $p_T$.

hep-ph↗
arXiv2022

Quarkonium polarization in Pb+Pb collisions in the improved color evaporation model

We apply our calculation of direct $J/\psi$ production in the improved color evaporation mode at ${\mathcal O}(\alpha_s^3)$ in collinear factorization to Pb+Pb collisions. The cold nuclear matter effects of nuclear modifications of the parton densities and intrinsic transverse momentum broadening are included. We find that the polarization is independent of these effects. Our calculations are compared to the Pb+Pb polarization parameters extracted by the ALICE Collaboration at $\sqrt{s_{NN}} = 5.02$ TeV.

hep-ph↗
arXiv2021

Production and polarization of direct $J/\psi$ to ${\mathcal O}(\alpha_s^3)$ in the improved color evaporation model in collinear factorization

We calculate the production and polarization of direct $J/\psi$ in the improved color evaporation model at $\mathcal{O}(\alpha_s^3)$ in the collinear factorization approach. We present the first calculation of polarization parameters $\lambda_\vartheta$, $\lambda_{\phi}$, and $\lambda_{\vartheta \phi}$ in the helicity and the Collins-Soper frames, as well as the frame-invariant polarization parameter $\tilde{\lambda}$ as a function of transverse momentum. We find agreement with both $J/\psi$ cross sections and the invariant polarization parameters at small and moderate $p_T$.

hep-ph↗
arXiv2018

Production and polarization of prompt $\varUpsilon$($n$S) in the improved color evaporation model using the $k_T$-factorization approach

We calculate the polarization of prompt $\varUpsilon$($n$S) production in the improved color evaporation model at leading order employing the $k_T$-factorization approach. We present the polarization parameter $\lambda_\vartheta$ of prompt $\varUpsilon$($n$S) as a function of transverse momentum in $p+p$ and $p+\bar{p}$ collisions to compare with data in the helicity, Collins-Soper and Gottfried-Jackson frames. We also present calculations of the bottomonium production cross sections as a function of transverse momentum and rapidity. This is the first $p_T$-dependent calculation of bottomonium production and polarization in the improved color evaporation model. We find agreement with both bottomonium cross sections and polarization measurements.

hep-ph↗
arXiv2018

Production and polarization of prompt $J/\psi$ in the improved color evaporation model using the $k_T$-factorization approach

We calculate the polarization of prompt $J/\psi$ production in the improved color evaporation model at leading order employing the $k_T$-factorization approach. In this paper, we present the polarization parameter $\lambda_\vartheta$ of prompt $J/\psi$ as a function of transverse momentum in $p+p$ and $p$ + A collisions to compare with data in the helicity, Collins-Soper and Gottfried-Jackson frames. We also present calculations of the charmonium production cross sections as a function of rapidity and transverse momentum. This is the first $p_T$-dependent calculation of charmonium polarization in the improved color evaporation model. We find agreement with both charmonium cross sections and polarization measurements.

hep-ph↗
arXiv2017

Polarization of prompt $J/\psi$ and $\Upsilon$(1S) production in the color evaporation model

We calculate the polarization of prompt $J/\psi$ and $\Upsilon$(1S) production using the color evaporation model at leading order. We present the polarization parameter $\lambda_\vartheta$ as a function of center of mass energy and rapidity in $p+p$ collisions. We also compare the $x_F$ dependence to experimental results in $p$+Cu and $\pi$+W collisions, and predict the $x_F$ dependence in $p$+Pb collisions at fixed-target energies. At energies far above the $Q\overline{Q}$ production threshold, we find the prompt $J/\psi$ and $\Upsilon$(1S) production to be longitudinally polarized with $\lambda_\vartheta^{J/\psi}=-0.51^{+0.05}_{-0.16}$ and $\lambda_\vartheta^{\Upsilon \rm{(1S)}}=-0.69^{+0.03}_{-0.02}$. Both prompt $J/\psi$ and prompt $\Upsilon$(1S) are also longitudinally polarized at central rapidity, becoming transversely polarized at the most forward rapidities.

hep-ph↗
arXiv2017

Polarized Heavy Quarkonium Production in the Color Evaporation Model

We explore polarized heavy quarkonium production using the color evaporation model at leading order. We present the polarized to total yield ratio as a function of center of mass energy and rapidity in $p+p$ collisions. At energies far above the $Q \overline Q$ production threshold, we find charmonium and bottomonium production to be longitudinally polarized ($J_z=0$). The quarkonium states are also longitudinally polarized at central rapidity, becoming transversely polarized ($J_z=\pm1$) at the most forward rapidities.

hep-ph↗
arXiv2012

Matrix Tile Analysis

Many tasks require finding groups of elements in a matrix of numbers, symbols or class likelihoods. One approach is to use efficient bi- or tri-linear factorization techniques including PCA, ICA, sparse matrix factorization and plaid analysis. These techniques are not appropriate when addition and multiplication of matrix elements are not sensibly defined. More directly, methods like bi-clustering can be used to classify matrix elements, but these methods make the overly-restrictive assumption that the class of each element is a function of a row class and a column class. We introduce a general computational problem, `matrix tile analysis' (MTA), which consists of decomposing a matrix into a set of non-overlapping tiles, each of which is defined by a subset of usually nonadjacent rows and columns. MTA does not require an algebra for combining tiles, but must search over discrete combinations of tile assignments. Exact MTA is a computationally intractable integer programming problem, but we describe an approximate iterative technique and a computationally efficient sum-product relaxation of the integer program. We compare the effectiveness of these methods to PCA and plaid on hundreds of randomly generated tasks. Using double-gene-knockout data, we show that MTA finds groups of interacting yeast genes that have biologically-related functions.

cs.LG↗