arXiv subjects

Harshit Pandey

Publications and source records attributed to Harshit Pandey.

12 recordsLinked to original sources

arXiv2026

Momentum Dependence of Heavy Quark Diffusion in a Thermal Gluonic Plasma on the Lattice

We study the dynamics of a heavy quark in a thermal plasma consisting of non-perturbatively interacting soft momentum gluons at high temperatures, described in terms of an effective theory of QCD. Discretizing this effective field theory on a three-dimensional lattice, we propose a numerical strategy that allows us to simulate the dynamics of a heavy quark for different values of initial momenta and for a wide temperature range, higher than $480$ MeV. This allows us, for the first time, to extract the momentum dependence of the heavy quark drag and diffusion coefficients in a non-perturbatively interacting thermal, non-Abelian plasma.

hep-lat↗

arXiv2026

Imprints of $U_A(1)$ chiral anomaly and disorder in the Dirac eigenspectrum of QCD at finite temperature

We perform a comprehensive study of the properties of Dirac eigenvalue spectrum in QCD as a function of temperature on the lattice. In addition to effects due to interplay between interactions and disorder inherently present in a many-body system, the Dirac spectrum also contains crucial information about the effective restoration of different subgroups of almost exact two-flavor chiral symmetry in QCD. We calculate the infrared eigenvalues of the overlap Dirac operator on 2+1 flavor QCD ensembles generated using domain wall fermion discretization, on a large volume lattice. From the normalized level spacing ratios, we identify those eigenvalues that have intermediate level statistics, distinctly different from the majority in the bulk spectrum that follow universal level fluctuations similar to a random matrix of Gaussian unitary type. We provide an explanation of these intermediate level ratios in terms of a specific random matrix model and quantify the correlation between these eigenstates and disorder in the gauge fields manifested in the renormalized Polyakov loop values. Whereas the existence of intermediate eigenmodes is intimately connected to the effective restoration of different subgroups of chiral symmetry close to chiral crossover transition, their origin can be traced to random uncorrelated disorder at higher temperatures when the $U_A(1)$ is effectively restored. We also, for the first time, calculate the Thouless conductance for the Dirac spectrum that quantifies the structural rigidity of the eigenvectors, and use it as a diagnostic tool to understand the restoration of the anomalous $U_A(1)$ subgroup of chiral symmetry and localization driven by disorder.

hep-lat↗

arXiv2025

Thermal SU(2) lattice gauge theory for intertwined orders and hole pockets in the cuprates

The cuprate pseudogap phase displays Fermi arc spectral weight in photoemission and scanning tunneling microscopy (STM), while recent magnetotransport observations yield evidence for the existence of hole pockets of fractional area $p/8$, where $p$ is the doping density. We present a Monte Carlo study of a thermal SU(2) lattice gauge theory which can reconcile these observations. Our simulation includes the SU(2) gauge field $U$ of a $\pi$-flux spin liquid, and a SU(2) fundamental charge $e$ Higgs boson $B$. There is a Yukawa coupling between $B$, the fermionic spinons of the spin liquid, and the hole pockets of a fractionalized Fermi liquid. At the higher temperatures of the pseudogap, the finite-doping sign problem is evaded by including only thermal fluctuations of $B$ and $U$, while the fermions are diagonalized exactly for each boson background. Our study also yields a fractionalized description of intertwined orders at lower temperatures, including the onset of $d$-wave superconductivity by the expulsion of vortices with flux $h/(2e)$, each with charge-order halos. We discuss conditions under which quantum oscillations in the density of states from hole pockets of area $p/8$ could be observable in clean under-hole-doped cuprates.

cond-mat.str-el↗

arXiv2025

Understanding thermalization in a non-Abelian gauge theory in terms of its soft modes

We measure the maximal Lyapunov exponent $\lambda_L$ of physical states in a SU(2) gauge theory consisting of soft momentum modes both in and out-of-thermal equilibrium conditions using ab-initio lattice techniques. We have implemented different algorithms to appropriately describe the dynamics of soft-modes for a wide range of temperatures and under non-equilibrium conditions. The non-equilibrium state has been realized starting from an over-occupied initial condition for low momentum soft gluons whereas the thermal state comprises of strongly interacting soft gluons at temperatures where these are well separated from the hard momentum modes. Spectra of positive Lyapunov exponents is observed in both these states, similar to a chaotic dynamical system. From the Kolmogorov-Sinai entropy rate measured in terms of this spectrum, we estimate a typical time-scale of $\sim 0.50(3)$ fm/c to achieve thermalization at $T\sim 600$ MeV starting from the non-thermal state. We also measure, for the first time, the $\lambda_L$ for long wavelength critical modes of SU(2) using the out-of-time-ordered correlator of a classical $Z_2$ scalar field theory, which shares the same universal behavior with SU(2), near the deconfinement phase transition. The $\lambda_L$ is observed to maximize at the transition temperature.

hep-lat↗

arXiv2024

Understanding the approach to thermalization from the eigenspectrum of non-Abelian gauge theories

We study some interesting aspects of the spectral properties of SU(3) gauge theory, both with and without dynamical quarks (QCD) at thermal equilibrium using lattice gauge theory techniques. By calculating the eigenstates of a massless overlap Dirac operator on the gauge configurations, we implement a gauge-invariant method to study spectral properties of non-Abelian gauge theories. We have unambiguously categorized Dirac eigenvalues into different regimes based on a quantity defined in terms of the ratios of nearest neighbor spacings. While majority of these eigenstates below the magnetic scale are similar to those of random matrices belonging to the Gaussian Unitary ensemble at temperatures much higher than the chiral crossover transition in QCD, a few among them start to become prominent only near the crossover. These form fractal-like clusters with the median value for their fractal dimensions hinting at the universality class of the chiral transition in QCD. We further demonstrate that momentum modes below the magnetic scale in a particular non-equilibrium state of QCD are classically chaotic and estimate an upper bound on the thermalization time $\sim 1.44$ fm/c by matching this magnetic scale with that of a thermal state at $\sim 600$ MeV.

hep-lat↗

arXiv2023

Heavy Quark Momentum Broadening in a Non-Abelian Plasma away from Thermal Equilibrium

We perform classical-statistical real-time lattice simulations to compute real-time spectral functions and momentum broadening of quarks in the presence of strongly populated non-Abelian gauge fields. Based on a novel methodology to extract the momentum broadening for relativistic quarks, we find that the momentum distribution of quarks exhibit interesting non-perturbative features as a function of time due to correlated momentum kicks it receives from the medium, eventually going over to a diffusive regime. We extract the momentum diffusion coefficient for a mass range describing charm and bottom quarks and find sizeable discrepancies from the heavy quark limit.

hep-lat↗

arXiv2021

Multitask Prompted Training Enables Zero-Shot Task Generalization

Large language models have recently been shown to attain reasonable zero-shot generalization on a diverse set of tasks (Brown et al., 2020). It has been hypothesized that this is a consequence of implicit multitask learning in language models' pretraining (Radford et al., 2019). Can zero-shot generalization instead be directly induced by explicit multitask learning? To test this question at scale, we develop a system for easily mapping any natural language tasks into a human-readable prompted form. We convert a large set of supervised datasets, each with multiple prompts with diverse wording. These prompted datasets allow for benchmarking the ability of a model to perform completely held-out tasks. We fine-tune a pretrained encoder-decoder model (Raffel et al., 2020; Lester et al., 2021) on this multitask mixture covering a wide variety of tasks. The model attains strong zero-shot performance on several standard datasets, often outperforming models up to 16x its size. Further, our approach attains strong performance on a subset of tasks from the BIG-bench benchmark, outperforming models up to 6x its size. All trained models are available at https://github.com/bigscience-workshop/t-zero and all prompts are available at https://github.com/bigscience-workshop/promptsource.

cs.LG↗

arXiv2021

DRIFT: A Toolkit for Diachronic Analysis of Scientific Literature

In this work, we present to the NLP community, and to the wider research community as a whole, an application for the diachronic analysis of research corpora. We open source an easy-to-use tool coined: DRIFT, which allows researchers to track research trends and development over the years. The analysis methods are collated from well-cited research works, with a few of our own methods added for good measure. Succinctly put, some of the analysis methods are: keyword extraction, word clouds, predicting declining/stagnant/growing trends using Productivity, tracking bi-grams using Acceleration plots, finding the Semantic Drift of words, tracking trends using similarity, etc. To demonstrate the utility and efficacy of our tool, we perform a case study on the cs.CL corpus of the arXiv repository and draw inferences from the analysis methods. The toolkit and the associated code are available here: https://github.com/rajaswa/DRIFT.

cs.CL↗

arXiv2021

Superpixel-based Knowledge Infusion in Deep Neural Networks for Image Classification

Superpixels are higher-order perceptual groups of pixels in an image, often carrying much more information than the raw pixels. There is an inherent relational structure to the relationship among different superpixels of an image such as adjacent superpixels are neighbours of each other. Our interest here is to treat these relative positions of various superpixels as relational information of an image. This relational information can convey higher-order spatial information about the image, such as the relationship between superpixels representing two eyes in an image of a cat. That is, two eyes are placed adjacent to each other in a straight line or the mouth is below the nose. Our motive in this paper is to assist computer vision models, specifically those based on Deep Neural Networks (DNNs), by incorporating this higher-order information from superpixels. We construct a hybrid model that leverages (a) Convolutional Neural Network (CNN) to deal with spatial information in an image and (b) Graph Neural Network (GNN) to deal with relational superpixel information in the image. The proposed model is learned using a generic hybrid loss function. Our experiments are extensive, and we evaluate the predictive performance of our proposed hybrid vision model on seven different image classification datasets from a variety of domains such as digit and object recognition, biometrics, medical imaging. The results demonstrate that the relational superpixel information processed by a GNN can improve the performance of a standard CNN-based vision system.

cs.CV↗

arXiv2021

LRG at TREC 2020: Document Ranking with XLNet-Based Models

Establishing a good information retrieval system in popular mediums of entertainment is a quickly growing area of investigation for companies and researchers alike. We delve into the domain of information retrieval for podcasts. In Spotify's Podcast Challenge, we are given a user's query with a description to find the most relevant short segment from the given dataset having all the podcasts. Previous techniques that include solely classical Information Retrieval (IR) techniques, perform poorly when descriptive queries are presented. On the other hand, models which exclusively rely on large neural networks tend to perform better. The downside to this technique is that a considerable amount of time and computing power are required to infer the result. We experiment with two hybrid models which first filter out the best podcasts based on user's query with a classical IR technique, and then perform re-ranking on the shortlisted documents based on the detailed description using a transformer-based model.

cs.IR↗

arXiv2021

LRG at SemEval-2021 Task 4: Improving Reading Comprehension with Abstract Words using Augmentation, Linguistic Features and Voting

In this article, we present our methodologies for SemEval-2021 Task-4: Reading Comprehension of Abstract Meaning. Given a fill-in-the-blank-type question and a corresponding context, the task is to predict the most suitable word from a list of 5 options. There are three sub-tasks within this task: Imperceptibility (subtask-I), Non-Specificity (subtask-II), and Intersection (subtask-III). We use encoders of transformers-based models pre-trained on the masked language modelling (MLM) task to build our Fill-in-the-blank (FitB) models. Moreover, to model imperceptibility, we define certain linguistic features, and to model non-specificity, we leverage information from hypernyms and hyponyms provided by a lexical database. Specifically, for non-specificity, we try out augmentation techniques, and other statistical techniques. We also propose variants, namely Chunk Voting and Max Context, to take care of input length restrictions for BERT, etc. Additionally, we perform a thorough ablation study, and use Integrated Gradients to explain our predictions on a few samples. Our best submissions achieve accuracies of 75.31% and 77.84%, on the test sets for subtask-I and subtask-II, respectively. For subtask-III, we achieve accuracies of 65.64% and 62.27%.

cs.CL↗

arXiv2021

NLRG at SemEval-2021 Task 5: Toxic Spans Detection Leveraging BERT-based Token Classification and Span Prediction Techniques

Toxicity detection of text has been a popular NLP task in the recent years. In SemEval-2021 Task-5 Toxic Spans Detection, the focus is on detecting toxic spans within passages. Most state-of-the-art span detection approaches employ various techniques, each of which can be broadly classified into Token Classification or Span Prediction approaches. In our paper, we explore simple versions of both of these approaches and their performance on the task. Specifically, we use BERT-based models -- BERT, RoBERTa, and SpanBERT for both approaches. We also combine these approaches and modify them to bring improvements for Toxic Spans prediction. To this end, we investigate results on four hybrid approaches -- Multi-Span, Span+Token, LSTM-CRF, and a combination of predicted offsets using union/intersection. Additionally, we perform a thorough ablative analysis and analyze our observed results. Our best submission -- a combination of SpanBERT Span Predictor and RoBERTa Token Classifier predictions -- achieves an F1 score of 0.6753 on the test set. Our best post-eval F1 score is 0.6895 on intersection of predicted offsets from top-3 RoBERTa Token Classification checkpoints. These approaches improve the performance by 3% on average than those of the shared baseline models -- RNNSL and SpaCy NER.

cs.CL↗