arXiv subjects

Jaspreet Singh

Publications and source records attributed to Jaspreet Singh.

At least 19 recordsLinked to original sources

arXiv2026

Exploring Student Perception on Gen AI Adoption in Higher Education: A Descriptive Study

The rapid proliferation of Generative Artificial Intelligence (GenAI) is reshaping pedagogical practices and assessment models in higher education. While institutional and educator perspectives on GenAI integration are increasingly documented, the student perspective remains comparatively underexplored. This study examines how students perceive, use, and evaluate GenAI within their academic practices, focusing on usage patterns, perceived benefits, and expectations for institutional support. Data were collected through a questionnaire administered to 436 postgraduate Computer Science students at the University of Hertfordshire and analysed using descriptive methods. The findings reveal a Confidence-Competence Paradox: although more than 60% of students report high familiarity with tools such as ChatGPT, daily academic use remains limited and confidence in effective application is only moderate. Students primarily employ GenAI for cognitive scaffolding tasks, including concept clarification and brainstorming, rather than fully automated content generation. At the same time, respondents express concerns regarding data privacy, reliability of AI-generated information, and the potential erosion of critical thinking skills. The results also indicate strong student support for integrating AI literacy into curricula and programme Knowledge, Skills, and Behaviours (KSBs). Overall, the study suggests that universities should move beyond a policing approach to GenAI and adopt a pedagogical framework that emphasises AI literacy, ethical guidance, and equitable access to AI tools.

cs.CY↗

arXiv2026

Accurate Shift Invariant Convolutional Neural Networks Using Gaussian-Hermite Moments

The convolutional neural networks (CNNs) are not inherently shift invariant or equivariant. The downsampling operation, used in CNNs, is one of the key reasons which breaks the shift invariant property of a CNN. Conversely, downsampling operation is important to improve computational efficiency and increase the area of the receptive field for more contextual information. In this work, we propose Gaussian-Hermite Sampling (GHS), a novel downsampling strategy designed to achieve accurate shift invariance. GHS leverages Gaussian-Hermite polynomials to perform shift-consistent sampling, enabling CNN layers to maintain invariance to arbitrary spatial shifts prior to training. When integrated into standard CNN architectures, the proposed method embeds shift invariance directly at the layer level without requiring architectural modifications or additional training procedures. We evaluate the proposed approach on CIFAR-10, CIFAR-100, and MNIST-rot datasets. Experimental results demonstrate that GHS significantly improves shift consistency, achieving 100% classification consistency under spatial shifts, while also improving classification accuracy compared to baseline CNN models.

cs.CV↗

arXiv2026

Interpretable Multimodal Gesture Recognition for Drone and Mobile Robot Teleoperation via Log-Likelihood Ratio Fusion

Human operators are still frequently exposed to hazardous environments such as disaster zones and industrial facilities, where intuitive and reliable teleoperation of mobile robots and Unmanned Aerial Vehicles (UAVs) is essential. In this context, hands-free teleoperation enhances operator mobility and situational awareness, thereby improving safety in hazardous environments. While vision-based gesture recognition has been explored as one method for hands-free teleoperation, its performance often deteriorates under occlusions, lighting variations, and cluttered backgrounds, limiting its applicability in real-world operations. To overcome these limitations, we propose a multimodal gesture recognition framework that integrates inertial data (accelerometer, gyroscope, and orientation) from Apple Watches on both wrists with capacitive sensing signals from custom gloves. We design a late fusion strategy based on the log-likelihood ratio (LLR), which not only enhances recognition performance but also provides interpretability by quantifying modality-specific contributions. To support this research, we introduce a new dataset of 20 distinct gestures inspired by aircraft marshalling signals, comprising synchronized RGB video, IMU, and capacitive sensor data. Experimental results demonstrate that our framework achieves performance comparable to a state-of-the-art vision-based baseline while significantly reducing computational cost, model size, and training time, making it well suited for real-time robot control. We therefore underscore the potential of sensor-based multimodal fusion as a robust and interpretable solution for gesture-driven mobile robot and drone teleoperation.

cs.RO↗

arXiv2025

Physisorption on Nanomechanical Resonators: The Overlooked Influence of Trace Moisture

Short gas pulses introduced in a vacuum chamber have long been utilized to showcase the ultra-low mass resolutions achievable with nanomechanical resonators. The resonance frequency shifts are used as evidence of gas adsorption. However, there is very little clarity as to what exactly is adsorbing on to the resonators. We demonstrate that the physisorption of gases on cantilevers is predominantly the effect of moisture content that is present even in ultra-high purity gases. The experimental work is performed at low temperatures and in a high vacuum and is supported by theoretical calculations and simulation.

cond-mat.mes-hall↗

arXiv2025

Local symmetry breaking and orbital glass behaviour in CoFe2O4

The structural distortions, orbital correlations, and electronic states in cobalt ferrite (CoFe2O4) were investigated using complementary characterisation techniques, including SR-XRD, HAXPES, XANES, EXAFS, and Raman spectroscopy. SR-XRD confirms phase purity and reveals a temperature-dependent superlattice reflection between 200 K and 100 K, consistent with the emergence of short-range orbital ordering driven by cooperative Jahn-Teller distortion (JTD). The disappearance of this feature below 100 K signals orbital freezing and the onset of a glass-like orbital state. HAXPES measurements show multiplet splitting and charge-transfer satellite features in the Co and Fe 2p core levels, indicating mixed valence states and strong electron correlations. XANES analysis reveals hybridized p-d states and local coordination distortions. Temperature-dependent EXAFS measurements indicate increasing local disorder-particularly in Fe-O and Fe-Fe octahedral bonds as evidenced by enhanced Debye-Waller factors. These distortions, attributed to cation redistribution and oxygen vacancies, are static and asymmetric, primarily affecting the octahedral sublattice. Notably, signatures of cooperative Jahn-Teller distortions emerge in the intermediate temperature range (200-100 K) and disappear upon further cooling. Raman spectroscopy further supports these findings, revealing phonon anomalies and enhanced spin-phonon coupling in the same temperature range. Magnetic measurements indicate spin reorientation and exchange interaction anomalies that align with the orbital behaviour. Together, these results hint at a frustrated orbital state in CoFe2O4 possibly involving cooperative Jahn-Teller distortions, disrupted long-range coherence, and orbital glass behaviour offering new insights into the coupling of orbital, spin, and lattice degrees of freedom in spinel systems.

cond-mat.str-el↗

arXiv2025

Exploring the Effect of Robotic Embodiment and Empathetic Tone of LLMs on Empathy Elicitation

This study investigates the elicitation of empathy toward a third party through interaction with social agents. Participants engaged with either a physical robot or a voice-enabled chatbot, both driven by a large language model (LLM) programmed to exhibit either an empathetic tone or remain neutral. The interaction is focused on a fictional character, Katie Banks, who is in a challenging situation and in need of financial donations. The willingness to help Katie, measured by the number of hours participants were willing to volunteer, along with their perceptions of the agent, were assessed for 60 participants. Results indicate that neither robotic embodiment nor empathetic tone significantly influenced participants' willingness to volunteer. While the LLM effectively simulated human empathy, fostering genuine empathetic responses in participants proved challenging.

cs.HC↗

arXiv2023

Data Augmentation for Sample Efficient and Robust Document Ranking

Contextual ranking models have delivered impressive performance improvements over classical models in the document ranking task. However, these highly over-parameterized models tend to be data-hungry and require large amounts of data even for fine-tuning. In this paper, we propose data-augmentation methods for effective and robust ranking performance. One of the key benefits of using data augmentation is in achieving sample efficiency or learning effectively when we have only a small amount of training data. We propose supervised and unsupervised data augmentation schemes by creating training data using parts of the relevant documents in the query-document pairs. We then adapt a family of contrastive losses for the document ranking task that can exploit the augmented data to learn an effective ranking model. Our extensive experiments on subsets of the MS MARCO and TREC-DL test sets show that data augmentation, along with the ranking-adapted contrastive losses, results in performance improvements under most dataset sizes. Apart from sample efficiency, we conclusively show that data augmentation results in robust models when transferred to out-of-domain benchmarks. Our performance improvements in in-domain and more prominently in out-of-domain benchmarks show that augmentation regularizes the ranking model and improves its robustness and generalization capability.

cs.IR↗

arXiv2023

Numerical Investigation of Water Entry of Hydrophobic Spheres

We perform numerical simulations to study the dynamics of the entry of hydrophobic spheres in a pool of water using ANSYS. To track the air-water interface during the translation of the sphere in the pool of water, we use the volume of fluid (VOF) model. The continuum surface force (CSF) method computes the surface tension force. To simulate the hydrophobic surface properties, we also include wall adhesion. We perform simulations with different diameters and impact speeds of the sphere. Our simulations capture the formation of different types of air cavities, pinch-offs of these cavities, and other finer details similar to the experiments performed at the same parameters. Finally, we compare the coefficient of drag among the different hydrophobic cases. We further perform simulations of hydrophilic spheres impacting the pool of water and compare the drag coefficient with the analogous hydrophobic cases. We conclude that the spheres with hydrophobic surfaces have a lower drag coefficient than their hydrophilic counterparts. This lower drag of the hydrophobic spheres is attributed to the formation of the air cavity by the hydrophobic surfaces while translating through the pool of water, which reduces the area of the sphere in contact with water. In contrast, no such air cavity forms in the case of hydrophilic spheres.

physics.flu-dyn↗

arXiv2023

Mechanical, Optical and Thermoelectric Properties of Janus BiTeCl Monolayer

We report mechanical, optical and thermoelectric properties of recently fabricated Janus BiTeCl monolayer using density functional and semi-classical Boltzmann transport theory. Janus BiTeCl monolayer exhibits a direct bandgap, high carrier mobility (~10$^3$ cm$^2$V$^{-1}$s$^{-1}$) and high optical absorption in the UV-visible region. The mechanical behavior of the Janus BiTeCl monolayer is nearly isotropic having an ideal tensile strength ~ 15 GPa. The higher value of the Gruneisen parameter ($γ$), a low value of phonon group velocity (vg), and very little phonon scattering time ($τ_p$) lead to low lattice thermal conductivity (1.46 W/mK) of Janus BiTeCl monolayer. The combined effect of thermal conductivity and electronic transport coefficients of Janus BiTeCl monolayer results in the figure of merit (ZT) in the range of 0.43-0.75 at 300-500 K. Our results suggest Janus BiTeCl monolayer be a potential candidate for optoelectronic and moderate temperature thermoelectric applications.

cond-mat.mtrl-sci↗

arXiv2023

First Principles Study of 2D Ring-Te and its Electrical Contact with Topological Dirac Semimetal

In recent years, researchers have manifested their interest in the two-dimensional (2D) mono-elemental materials of group-VI elements because of their excellent optoelectronic, photovoltaic and thermoelectric properties. Despite the intensive recent research efforts, there is still a possibility of novel 2D allotropes of these elements due to their multivalency nature. Here, we have predicted a novel 2D allotrope of tellurium (ring-Te) using density functional theory. Its stability is confirmed by phonon and ab-initio molecular dynamics simulations. The ring-Te has an indirect band gap of 0.69 eV (1.16 eV) at PBE (HSE06) level of theories and undergoes an indirect-direct band gap transition under the tensile strain. The higher carrier mobility of holes (~103cm$^2$V$^{-1}$s$^{-1}$), good UV-visible light absorption ability and low exciton binding (~0.35 eV) of ring-Te gives rise to its potential applications in optoelectronic devices. Further, the electrical contact of ring-Te with topological Dirac semimetal (sq-Te) under the influence of electric field shows that the Schottky barriers and contact types can undergo transition from p-type to n-type Schottky contact and then to ohmic contact at higher electric field. Our study provides an insight into the physics of designing high-performance electrical coupled devices composed of 2D semiconductors and topological semimetals.

cond-mat.mtrl-sci↗

arXiv2023

Janus $β$-Te$_2$X (X = S, Se) Monolayers for Efficient Excitonic Solar Cells and Photocatalytic Water Splitting

Highly efficient, environmental friendly and renewable sources of energy are of great need today to combat with increasing energy demands and environmental pollution. In this work, we have investigated the novel 2D allotropes i.e., $β$-Te$_2$X (X = S, Se) using first-principles calculations and study their potential applications in light harvesting devices. Both the monolayers possess to have the high stability and semiconducting nature with an indirect band gap. The high carrier mobilities and excellent optical absorption of these monolayers make them potential candidates for solar conversion applications. We have proposed the type-II heterojunction solar cells and calculated their power conversion efficiencies (PCEs). The small conduction band offset and appropriate band gap of donor material in case of $β$-Te$_2$S(S-Side)/$α$-Te$_2$S(Te-Side) heterojunction results in the PCE of ~ 21%. In addition to that, the band alignments of these monolayers properly engulf the redox potentials of the water. The overpotentials required to trigger the hydrogen reduction (HER) and water oxidation (OER) half reactions reveal that HER and OER preferred the acidic and neutral mediums, respectively. The calculated solar-to-hydrogen (STH) efficiencies of $β$-Te$_2$S ($β$-Te$_2$Se) monolayers come out to be ~ 13 % (~12 %), respectively, which implies their practical applications in water splitting. Thus, our work provides strong evidence regarding the potential applications of these materials in the field of light harvesting devices.

cond-mat.mtrl-sci↗

arXiv2023

As-based ternary Janus monolayers for efficient thermoelectric and photocatalytic applications

Highly efficient and sustainable resources of energy are of great demand today to combat with environmental pollution and the energy crisis. In this work, we have examined the novel 2D Janus AsTeX (X = Cl, Br and I) monolayers using first-principles calculations and explore their potential energy conversion applications. We have demonstrated the thermal, energetic, dynamic and mechanical stability of AsTeX (X = Cl, Br, and I) monolayers. Janus AsTeX (X = Cl, Br and I) monolayers are indirect bandgap semiconductors with high carrier mobilities and excellent visible light optical absorption. Our findings demonstrate that the Janus AsTeCl and AsTeBr monolayers exhibits low lattice thermal conductivity and excellent electronic transport properties obtained using semi-classical Boltzmann transport theory including various scattering mechanism. Additionally, the redox potential of water is adequately engulfed by the band alignments of the AsTeCl and AsTeBr monolayers. The water splitting process under illumination can proceeds spontaneously on Janus AsTeBr monolayer, while a minimal low external potential (0.26-0.29 eV) is required to trigger water splitting process on Janus AsTeCl monolayer. A more than 10% STH efficiency of these monolayers indicate their potential practical applications in the commercial production of hydrogen. Thus, our study demonstrates that these monolayers can show potential applications in energy conversion fields.

cond-mat.mtrl-sci↗

arXiv2023

Earthquake Magnitude and b value prediction model using Extreme Learning Machine

Earthquake prediction has been a challenging research area for many decades, where the future occurrence of this highly uncertain calamity is predicted. In this paper, several parametric and non-parametric features were calculated, where the non-parametric features were calculated using the parametric features. $8$ seismic features were calculated using Gutenberg-Richter law, the total recurrence, and the seismic energy release. Additionally, criterions such as Maximum Relevance and Maximum Redundancy were applied to choose the pertinent features. These features along with others were used as input for an Extreme Learning Machine (ELM) Regression Model. Magnitude and time data of $5$ decades from the Assam-Guwahati region were used to create this model for magnitude prediction. The Testing Accuracy and Testing Speed were computed taking the Root Mean Squared Error (RMSE) as the parameter for evaluating the mode. As confirmed by the results, ELM shows better scalability with much faster training and testing speed (up to a thousand times faster) than traditional Support Vector Machines. The testing RMSE came out to be around $0.097$. To further test the model's robustness -- magnitude-time data from California was used to calculate the seismic indicators which were then fed into an ELM and then tested on the Assam-Guwahati region. The model proves to be robust and can be implemented in early warning systems as it continues to be a major part of Disaster Response and management.

physics.geo-ph↗

arXiv2022

Learning Invariant Representations for Equivariant Neural Networks Using Orthogonal Moments

The convolutional layers of standard convolutional neural networks (CNNs) are equivariant to translation. However, the convolution and fully-connected layers are not equivariant or invariant to other affine geometric transformations. Recently, a new class of CNNs is proposed in which the conventional layers of CNNs are replaced with equivariant convolution, pooling, and batch-normalization layers. The final classification layer in equivariant neural networks is invariant to different affine geometric transformations such as rotation, reflection and translation, and the scalar value is obtained by either eliminating the spatial dimensions of filter responses using convolution and down-sampling throughout the network or average is taken over the filter responses. In this work, we propose to integrate the orthogonal moments which gives the high-order statistics of the function as an effective means for encoding global invariance with respect to rotation, reflection and translation in fully-connected layers. As a result, the intermediate layers of the network become equivariant while the classification layer becomes invariant. The most widely used Zernike, pseudo-Zernike and orthogonal Fourier-Mellin moments are considered for this purpose. The effectiveness of the proposed work is evaluated by integrating the invariant transition and fully-connected layer in the architecture of group-equivariant CNNs (G-CNNs) on rotated MNIST and CIFAR10 datasets.

cs.CV↗

arXiv2021

The Arcanum Mission: Scientific Objectives and Instruments for Neptune, Triton and KBOs

The Arcanum mission is a proposed L-class spacecraft that highlights the revolutionary approach which can now be taken to future space mission design. Using the case of the SpaceX Starship vehicle and in particular the high mass and volume characteristics of this launcher, the feasible large size of future missions, even with high delta-V transfer requirements, are analysed. A demonstrator vehicle, designed to support a large and capable science platform with multiple components, is detailed, clearly showing the range and depth of science goals that will be answerable thanks to the current revolution in super heavy-lift launch vehicles.

astro-ph.IM↗

arXiv2021

BERTnesia: Investigating the capture and forgetting of knowledge in BERT

Probing complex language models has recently revealed several insights into linguistic and semantic patterns found in the learned representations. In this paper, we probe BERT specifically to understand and measure the relational knowledge it captures. We utilize knowledge base completion tasks to probe every layer of pre-trained as well as fine-tuned BERT (ranking, question answering, NER). Our findings show that knowledge is not just contained in BERT's final layers. Intermediate layers contribute a significant amount (17-60%) to the total knowledge found. Probing intermediate layers also reveals how different types of knowledge emerge at varying rates. When BERT is fine-tuned, relational knowledge is forgotten but the extent of forgetting is impacted by the fine-tuning objective but not the size of the dataset. We found that ranking models forget the least and retain more knowledge in their final layer. We release our code on github to repeat the experiments.

cs.CL↗

arXiv2021

BERTnesia: Investigating the capture and forgetting of knowledge in BERT

Probing complex language models has recently revealed several insights into linguistic and semantic patterns found in the learned representations. In this article, we probe BERT specifically to understand and measure the relational knowledge it captures in its parametric memory. While probing for linguistic understanding is commonly applied to all layers of BERT as well as fine-tuned models, this has not been done for factual knowledge. We utilize existing knowledge base completion tasks (LAMA) to probe every layer of pre-trained as well as fine-tuned BERT models(ranking, question answering, NER). Our findings show that knowledge is not just contained in BERT's final layers. Intermediate layers contribute a significant amount (17-60%) to the total knowledge found. Probing intermediate layers also reveals how different types of knowledge emerge at varying rates. When BERT is fine-tuned, relational knowledge is forgotten. The extent of forgetting is impacted by the fine-tuning objective and the training data. We found that ranking models forget the least and retain more knowledge in their final layer compared to masked language modeling and question-answering. However, masked language modeling performed the best at acquiring new knowledge from the training data. When it comes to learning facts, we found that capacity and fact density are key factors. We hope this initial work will spur further research into understanding the parametric memory of language models and the effect of training objectives on factual knowledge. The code to repeat the experiments is publicly available on GitHub.

cs.CL↗

arXiv2021

Towards Axiomatic Explanations for Neural Ranking Models

Recently, neural networks have been successfully employed to improve upon state-of-the-art performance in ad-hoc retrieval tasks via machine-learned ranking functions. While neural retrieval models grow in complexity and impact, little is understood about their correspondence with well-studied IR principles. Recent work on interpretability in machine learning has provided tools and techniques to understand neural models in general, yet there has been little progress towards explaining ranking models. We investigate whether one can explain the behavior of neural ranking models in terms of their congruence with well understood principles of document ranking by using established theories from axiomatic IR. Axiomatic analysis of information retrieval models has formalized a set of constraints on ranking decisions that reasonable retrieval models should fulfill. We operationalize this axiomatic thinking to reproduce rankings based on combinations of elementary constraints. This allows us to investigate to what extent the ranking decisions of neural rankers can be explained in terms of retrieval axioms, and which axioms apply in which situations. Our experimental study considers a comprehensive set of axioms over several representative neural rankers. While the existing axioms can already explain the particularly confident ranking decisions rather well, future work should extend the axiom set to also cover the other still "unexplainable" neural IR rank decisions.

cs.IR↗