arXiv subjects

Akanksha Singh

Publications and source records attributed to Akanksha Singh.

15 recordsLinked to original sources

arXiv2026

Selective Capability Unlearning in End-to-End Spoken Language Understanding

Modern spoken language understanding (SLU) systems are increasingly deployed in real-world settings, where specific functionalities may need to be removed due to policy or safety constraints. In SLU, a functionality corresponds to an intent and its associated slot-generation behavior. However, in autoregressive models, suppressing a target intent does not eliminate the conditional mapping that generates slots conditioned on that intent. When the intent prefix is externally supplied, the model can reconstruct the original intent-slot structure. We identify this structural failure as \textbf{\emph{capability persistence}}. We propose \textit{\underline{B}inding \underline{S}ubspace (BSU)}, a representation-level framework that isolates and attenuates intent-conditioned directions underlying this mapping. Across SLU benchmarks, BSU substantially reduces forced-prefix recoverability while preserving retained performance.

cs.CL↗

arXiv2025

H-QuEST: Accelerating Query-by-Example Spoken Term Detection with Hierarchical Indexing

Query-by-example spoken term detection (QbE-STD) searches for matching words or phrases in an audio dataset using a sample spoken query. When annotated data is limited or unavailable, QbE-STD is often done using template matching methods like dynamic time warping (DTW), which are computationally expensive and do not scale well. To address this, we propose H-QuEST (Hierarchical Query-by-Example Spoken Term Detection), a novel framework that accelerates spoken term retrieval by utilizing Term Frequency and Inverse Document Frequency (TF-IDF)-based sparse representations obtained through advanced audio representation learning techniques and Hierarchical Navigable Small World (HNSW) indexing with further refinement. Experimental results show that H-QuEST delivers substantial improvements in retrieval speed without sacrificing accuracy compared to existing methods.

eess.AS↗

arXiv2025

Cosmic reverberations on a constrained $ f(Q,T) $-model of the Universe

In this paper, we construct an isotropic cosmological model in the $ f(Q, T) $ theory of gravity in the frame of a flat FLRW spacetime being $ Q $ the non-metricity tensor and $ T $ the trace of the energy-momentum tensor. The gravity function is taken to be a quadratic equation, $ f(Q, T)=ζQ^2 + γT $, where $ ζ<0 $ and $ γ$ are the arbitrary constants. We constrain the model parameters $ α$ and $ H_0 $ using the recent observational datasets: the Hubble dataset (OHD), the $ Pantheon $ dataset of $ 1048 $ points, and the joint dataset (OHD + $ Pantheon $). The universe model transitions from an early deceleration state to an acceleration in late times. This model also provides the ekpyrotic phase of the universe on the redshift $ z>12.32 $. In this model, the Big Bang is described as a collision of branes, and thus, the Big Bang is not the beginning of time. Before the Big Bang, there is an ekpyrotic phase with the equation of state $ ω>> 1 $. In late times, the undeviating Hubble measurements reduce the $ H_0 $ tension in the reconstructed $ f(Q, T) $ function. Additionally, we study various physical parameters of the model. Finally, our model describes a quintessence dark energy model at later times.

gr-qc↗

arXiv2024

Evolution of the universe with quintessence model in Rastall gravity

We investigate the universe's evolution within the framework of Rastall gravity, which is an extension of the standard $Λ$CDM model. Utilizing a linear parametrization of the Equation of State (EoS) in a Friedmann-Lema\^ıtre-Robertson-Walker (FLRW) background, we constrain the model parameters through analysis of cosmic chronometers (CC), Pantheon, Gold, Gamma Ray Burst (GRB), and Baryon Acoustic Oscillations (BAO) datasets, as well as their joint analysis, under $1σ$ and $2σ$ confidence levels, considering the Rastall parameter $λ$. The constrained parameters are then used to compare our model with the standard $Λ$CDM model. Our findings include a detailed examination of the model's physical interpretations and demonstrate the potential for an accelerating universe expansion in later times, aligning with the observed behavior of dark energy.

gr-qc↗

arXiv2024

The consequence of higher-order curvature-based constraints on $ f(R, L_m) $ gravity

In this investigation, we perform an observational statistical analysis in the theory of $ f(R, L_m) $ gravity. The proposed theoretical model is based on the Ricci scalar's non-linear contribution. We use a distinct parameterization for the deceleration parameter and constrain the model parameters by using various observational data. To determine the best-fit model for the cosmological parameters, we use different observational datasets such as the Hubble Space Telescope, the Pantheon Supernova Survey, the Gold dataset, the Gamma-Ray Burst (GRB), and the Baryon Acoustic Oscillations (BAO). Furthermore, we study the late-time cosmic evolution of the Universe in detail and examine the implications of the constraint values on cosmological parameters. Additionally, we conduct a thorough comparison with the standard cosmological model $ Λ$CDM and other standard models obtained by Odintsov et al. \cite{Odintsov:2023cli, Odintsov:2024lid} to examine the validity of our proposed model in the low-redshift regimes. Finally, we find that the proposed model encapsulates an intriguing transition from early deceleration at high redshift to acceleration at low redshift, a quintessence dark energy scenario, and convergence towards the well-established $ Λ$CDM model in late-time Universe's evolution.

gr-qc↗

arXiv2024

Stability analysis of a dark energy model in Rastall gravity

We study a cosmological model in Rastall's theory of gravity in the framework of the flat FLRW metric. We formulate the value of the Hubble parameter, which contains two model parameters, $ α$ and $ j $. Employing the Markov Chain Monte Carlo (MCMC) sampling technique, we determine the values of these model parameters along with their uncertainties. Moreover, we derive the equation of state (EoS) parameter, which converges around the quintessence region. We perform a dynamical system analysis using the linearization technique to validate the results independently. Also, we discuss various physical attributes of the model, highlighting the transition to acceleration and the violation of the strong energy condition (SEC) in the late stages of evolution. In conclusion, our model mimics the behavior of a dark matter fluid during the past epoch and transitions into a quintessence dark energy model in the future epoch.

gr-qc↗

arXiv2024

Late time phantom characteristic of the model in $f(R,T)$ gravity with quadratic curvature term

We propose a novel cosmological framework within the $f(R,T)$ type modified gravity theory, incorporating a non-minimally coupled with the higher order of the Ricci scalar ($R$) as well as the trace of the energy-momentum tensor ($T$). Therefore, our well-motivated chosen $f(R,T)$ expression is $ R + R^m + 2 λT^n$, where $λ$, $m$, and $n$ are arbitrary constants. Taking a constant jerk parameter ($j$), we derive expressions for the deceleration parameter ($q$) and the Hubble parameter ($H$) as functions of the redshift $z$. We constrained our model with the recent Observational Hubble Dataset (OHD), $Pantheon$, and $ Pantheon $ + OHD datasets by using the analysis of Markov Chain Monte Carlo (MCMC). Our model shows early deceleration followed by late-time acceleration, with the transition occurring in the redshift range $1.10 \leq z_{tr} \leq 1.15$. Our findings suggest that this higher-order model of $f(R,T)$ gravity theory can efficiently provide a dark energy model for addressing the current scenario of cosmic acceleration.

gr-qc↗

arXiv2023

Comparison of fractional-order generalized wavelets and orthonormal wavelets methods for solving multi-dimensional fractional optimal control problems

This paper presents an efficient numerical technique for solving multi-dimensional fractional optimal control problems using fractional-order generalized Bernoulli wavelets. The numerical results obtained by this method have been compared with the results obtained by the method using orthonormal Bernoulli wavelets. Using fractional-order generalized Bernoulli wavelets, product and integral operational matrices have been obtained. By using these operational matrices, the multi-dimensional fractional optimal control problems have been reduced into a system of algebraic equations. To confirm the efficiency of the proposed numerical technique involving fractional-order generalized Bernoulli wavelets, numerical problems have been solved by using both orthonormal Bernoulli wavelets and fractional-order generalized Bernoulli wavelets and obtaining an approximate cost function value, which has been found by approximating state and control functions. In addition, the convergence rate and error bound of the proposed numerical method have also been derived.

math.OC↗

arXiv2023

A numerical technique for solving multi-dimensional fractional optimal control problems using fractional wavelet method

This paper presents an efficient numerical method for solving fractional optimal control problems using an operational matrix for a fractional wavelet. Using well-known formulae such as Caputo and Riemann-Liouville operators to determine fractional derivatives and integral fractional wavelets, operational matrices were devised and utilised to solve fractional optimal control problems. The proposed method reduced the fractional optimal control problems into a system of algebraic equations. To validate the effectiveness of the presented numerical approach, some illustrative problems were solved using fractional Taylor and Taylor wavelets, and the approximate cost function value derived by approximating state and control functions was compared. In addition, convergence rate and error bound of the proposed method have been derived.

math.OC↗

arXiv2023

A non-singular bouncing cosmology in $ f(R,T) $ gravity

We investigate a bounce realization in the framework of higher order curvature in $ f(R,T) $ modified theory of gravity. We perform a detailed analysis of the cosmological parameters to explain the contraction phase, the bounce phase, and the expansion phase. Furthermore, we observe a violation of the null energy condition, instability of the model, and a singularity upon deceleration at the bouncing point, which are the supporting results for a bouncing cosmology. The outcome of the slow roll parameters is satisfactory to understand the inflation era and the equation of state parameter exhibits a ghost condensate behavior of the model near the bounce. Additionally, we discuss the stability of the model using linear perturbations in the Hubble parameter as well as the energy density.

gr-qc↗

arXiv2022

The cosmological model in $ f(R,T^ϕ) $ gravity with Scalar Field conformity

The homogeneous and isotropic cosmological model in generalized $ f(R,T^ϕ) $ theories associated with scalar field is discussed, which is motivated by the $ f(R,T) $ theory of gravity studied by Harko et al. \cite{Harko:2011kv, Harko:2014pqa}. The $ f(R,T^ϕ) $ gravity can be explained as $ f(R,T) $ gravity with a self-interacting scalar field $ ϕ$, where $ T^ϕ$ is the trace of the energy-momentum tensor. The parametrization of Hubble parameter $ H(t) $ is taken as $ α-βe^{-γt} $, where $ α$, $β$ and $γ$ are arbitrary constants such that $ α, γ>0 $ and $ β<0 $. The model shows no space-time singularity and the expansion of the universe will continue forever, i.e., the future scenario of the universe attains Big Freeze. The model predicts the moderate inflationary scenario at the time of the evolution of the universe and it is consistent with $ Λ$CDM in late times. The consistency of the model has also been examined using recent observational Hubble dataset and supernovae dataset. Finally, the physical features of the model have been discussed in some detail.

gr-qc↗

arXiv2022

Sizing and Location Selection of Medium-Voltage Back-to-Back Converter for DER-Dominated Distribution Systems

Medium-voltage back-to-back (MVB2B) converters can connect two distribution systems and quantifiably transfer power between them. This function can enable the MVB2B converter to exchange distributed energy resource (DER)-generated power between two systems and bring significant value to enhancing distribution system DER adoption. Our previous work analyzed and demonstrated the value the MVB2B converter can bring to DER integration. As continuous work, this paper presents a methodology that helps address the MVB2B converter sizing and location selection problem in distribution systems with high DER penetrations. The proposed methodology aims to address three critical problems for MVB2B converter implementation in the real world: 1) which distribution systems are better to be connected, 2) what converter size is appropriate for connecting the distribution systems, and 3) where the optimal connection points are in the systems for connecting the MVB2B converter. The proposed methodology has been demonstrated by case studies that include various scenarios involving distribution systems with different dominated load types and high photovoltaic penetrations.

eess.SY↗

arXiv2022

Dynamics of a parametrized dark energy model in $f(R,T)$ gravity

We investigate a flat FLRW-model in $f(R,T)$-gravity, which includes the quadratic variation in scalar curvature $R$ and the linear term of the trace of the stress-energy tensor $T$. In turn, we establish the model has the behaviour of the late time Universe, which is accelerated expanding and ends up in a big rip. Using the parametrization of scale factor $a(t)$, we propose a model, which begins with point-type singularity, i.e., the model starts with a point of zero volume, infinite energy density and infinite temperature. The model's behaviour is accelerated expanding at present and $Λ$CDM in late times. Finally, the proposed model behaves like a quintessence dark energy model in the present time and is consistent with standard cosmology $Λ$CDM in late times.

gr-qc↗

arXiv2022

Theory, phenomenology, and experimental avenues for dark showers: a Snowmass 2021 report

In this work, we consider the case of a strongly coupled dark/hidden sector, which extends the Standard Model (SM) by adding an additional non-Abelian gauge group. These extensions generally contain matter fields, much like the SM quarks, and gauge fields similar to the SM gluons. We focus on the exploration of such sectors where the dark particles are produced at the LHC through a portal and undergo rapid hadronization within the dark sector before decaying back, at least in part and potentially with sizeable lifetimes, to SM particles, giving a range of possibly spectacular signatures such as emerging or semi-visible jets. Other, non-QCD-like scenarios leading to soft unclustered energy patterns or glueballs are also discussed. After a review of the theory, existing benchmarks and constraints, this work addresses how to build consistent benchmarks from the underlying physical parameters and present new developments for the PYTHIA Hidden Valley module, along with jet substructure studies. Finally, a series of improved search strategies is presented in order to pave the way for a better exploration of the dark showers at the LHC.

hep-ph↗

arXiv2022

Grid Value Analysis of Medium Voltage Back-to-Back Converter on DER Hosting Enhancement

This paper presents an analysis of the value that can be realized by medium-voltage back-to-back (MVB2B) converters in terms of increased utilization rate of distributed energy resource (DER) and the improvement in operational conditions. A systematic, transferrable, and scalable methodology has been designed to analyze and quantify the increased DER value from three perspectives: 1) curtailment reduction of the DER generation, 2) size reduction of the energy storage needed to otherwise realize DER hosting levels, and 3) hosting capacity improvement of DER compared to base distribution circuit capability. In the case study, the proposed methodology is applied to two utility distribution systems for analysis and quantification of the grid value of the MVB2B converter, installed in the distribution circuit, and provided to the solar photovoltaic (PV) DERs. The analysis results demonstrate that the MVB2B converter can deliver significant value to PV hosting enhancement of two adjacent distribution systems when they are connected by the MVB2B converter. Based on this case study, this paper analyzes and summarizes the approximate realized grid value of the MVB2B converter for distribution systems dominated by different shares of customer classes.

eess.SY↗