Pranjal Kumar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Pranjal Kumar has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Siddhartha Chauhan
Department of Computer Science and Engineering, National Institute of Technology, Hamirpur, Hamirpur, India
5 shared publications Lalit Kumar Awasthi
National Institute of Technology Hamirpur, Hamirpur, India
3 shared publicationsSiddhartha Chauhan
National Institute of Technology Hamirpur, Hamirpur, India
3 shared publications- Gurneet Singh
Lovely Professional University
2 shared publications - Piyush Rawat
Department of Systemics, School of Computer Science, University of Petroleum and Energy Studies, Dehradun, India
2 shared publications - Arun Raj Kumar Parthiban
National Institute of Technology Calicut
1 shared publication
Latest Publications
- Open Access
- 11 June 2026
- Published by MDPI AG in Preprints.org
Quantum SVMs with LoRA-Adapted Embeddings and Variational Circuits for Image Classification
The expressivity and scalability of fixed feature embeddings and shallow quantum circuits for Quantum Support Vector Machines (QSVMs) are restricted due to the limitations of near-term hardware. In this work, we propose a novel QSVM framework consisting of Low-Rank Adaptation (LoRA)-modified VisionTransformer embeddings as well as trainable variational quantum circuits. The architecture enhances the representational power by the joint optimization of embedding and circuit parameters. Empirical evaluation on distilled benchmark datasets shows significant performance gains, with up to 38.7% absolute accuracy improvement on MNIST (61.3% to 100.0%). The model also achieves 91% to 100% test accuracies on Fashion-MNIST, KMNIST and CIFAR-10 datasets with only 200 training samples for each dataset. The results show that the performance improvements are due to the joint design between the feature embeddings and the quantum circuits, rather than just the quantum kernels. Cross-dataset evaluations and circuit-depth analyses provide experimental evidence that the approach is suitable for scalable quantum machine learning in image classification tasks.
- 11 June 2026
- Published by MDPI AG in Preprints.org
Open AccessQuantum SVMs with LoRA-Adapted Embeddings and Variational Circuits for Image Classification
The expressivity and scalability of fixed feature embeddings and shallow quantum circuits for Quantum Support Vector Machines (QSVMs) are restricted due to the limitations of near-term hardware. In this work, we propose a novel QSVM framework consisting of Low-Rank Adaptation (LoRA)-modified VisionTransformer embeddings as well as trainable variational quantum circuits. The architecture enhances the representational power by the joint optimization of embedding and circuit parameters. Empirical evaluation on distilled benchmark datasets shows significant performance gains, with up to 38.7% absolute accuracy improvement on MNIST (61.3% to 100.0%). The model also achieves 91% to 100% test accuracies on Fashion-MNIST, KMNIST and CIFAR-10 datasets with only 200 training samples for each dataset. The results show that the performance improvements are due to the joint design between the feature embeddings and the quantum circuits, rather than just the quantum kernels. Cross-dataset evaluations and circuit-depth analyses provide experimental evidence that the approach is suitable for scalable quantum machine learning in image classification tasks.
- 29 May 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 260
Real-time change-point detection and impact quantification for smart-grid PMU cybersecurity
- 29 May 2026
- Published by Elsevier in Electric Power Systems Research
- Vol. 260
Real-time change-point detection and impact quantification for smart-grid PMU cybersecurity
- 9 April 2026
- Published by Springer Nature in Intelligent Service Robotics
- Vol. 19 (3) , 54
Temporal modeling for domain-invariant fault diagnosis in robotics digital twin systems
- 9 April 2026
- Published by Springer Nature in Intelligent Service Robotics
- Vol. 19 (3) , 54
Temporal modeling for domain-invariant fault diagnosis in robotics digital twin systems
- 24 March 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 174
Robust geometry-adaptive transformer engine for simulation-to-real point cloud classification
- 24 March 2026
- Published by Elsevier in Engineering Applications of Artificial Intelligence
- Vol. 174
Robust geometry-adaptive transformer engine for simulation-to-real point cloud classification
- 2 January 2026
- Published by Springer Nature
- p. 301-311
Fine-Tuning and Efficient Quantization for Optimization of Diffusion Models
- 2 January 2026
- Published by Springer Nature
- p. 301-311
Fine-Tuning and Efficient Quantization for Optimization of Diffusion Models
- Open Access
- 28 July 2025
- Published by MDPI AG in Preprints.org
Temporal Modeling for Domain-Invariant Fault Diagnosis in Robotics Digital Twin Systems
Fault diagnosis in robotics frequently encounters obstacles due to the scarcity of labeled real-world data and the disparities between simulated and physical systems. Digital twin technology mitigates data limitations by producing simulated datasets; however, the sim-to-real gap remains a barrier to effective model generalization. This study introduces the TemporalTwinNet (TTN) framework, a novel approach designed for digital twin-supported fault diagnosis in robotics. By embedding bidirectional Long Short-Term Memory (LSTM) layers into the feature extraction process, the TTN framework adeptly captures temporal dependencies within time-series data, enhancing the alignment of simulated and real-world trajectories. Tested on an open-source robotics dataset comprising 3,600 simulated and 90 real samples, the proposed TTN achieves a real-world test precision of 86.67\%, significantly narrowing the sim-to-real gap to 9.44\%. Notably, the model elevates F1 scores for difficult categories, such as the healthy state (improving from 0.06 to 0.63), while sustaining strong simulation performance with a precision of 96.11\%. Additionally, the framework integrates severity prediction, boosting its practical utility. These outcomes underscore the efficacy of temporal modeling in overcoming the sim-to-real divide, presenting a resilient solution for predictive maintenance in robotics.
- 28 July 2025
- Published by MDPI AG in Preprints.org
Open AccessTemporal Modeling for Domain-Invariant Fault Diagnosis in Robotics Digital Twin Systems
Fault diagnosis in robotics frequently encounters obstacles due to the scarcity of labeled real-world data and the disparities between simulated and physical systems. Digital twin technology mitigates data limitations by producing simulated datasets; however, the sim-to-real gap remains a barrier to effective model generalization. This study introduces the TemporalTwinNet (TTN) framework, a novel approach designed for digital twin-supported fault diagnosis in robotics. By embedding bidirectional Long Short-Term Memory (LSTM) layers into the feature extraction process, the TTN framework adeptly captures temporal dependencies within time-series data, enhancing the alignment of simulated and real-world trajectories. Tested on an open-source robotics dataset comprising 3,600 simulated and 90 real samples, the proposed TTN achieves a real-world test precision of 86.67\%, significantly narrowing the sim-to-real gap to 9.44\%. Notably, the model elevates F1 scores for difficult categories, such as the healthy state (improving from 0.06 to 0.63), while sustaining strong simulation performance with a precision of 96.11\%. Additionally, the framework integrates severity prediction, boosting its practical utility. These outcomes underscore the efficacy of temporal modeling in overcoming the sim-to-real divide, presenting a resilient solution for predictive maintenance in robotics.
- 2 April 2025
- Published by Springer Nature in Archives of Computational Methods in Engineering
- Vol. 32 (7) , 4049-4092
Diffusion Models and Generative Artificial Intelligence: Frameworks, Applications and Challenges
- 2 April 2025
- Published by Springer Nature in Archives of Computational Methods in Engineering
- Vol. 32 (7) , 4049-4092
Diffusion Models and Generative Artificial Intelligence: Frameworks, Applications and Challenges
- Open Access
- 4 February 2025
- Published by MDPI AG in Preprints.org
Fine Tuning and Efficient Quantization for Optimization of Diffusion Models
Diffusion models have emerged as a critical component in the domain of deep generative models for image synthesis. However, the substantial computational demands of their practical deployment limit the viability of developing next-generation machine learning algorithms. This paper presents an optimization strategy that includes quantization, fine-tuning, and inference techniques to improve the effectiveness of diffusion models. It also discusses identifying and reducing particular training bottlenecks. By means of thorough testing and assessment, the suggested enhancements are methodically compared with baseline models. While preserving similar image quality, the optimization methods show improved computational efficiency. This advancement makes it easier to create diffusion models that are more precise and scalable, enabling wider uses in computer vision and related domains.
- 4 February 2025
- Published by MDPI AG in Preprints.org
Open AccessFine Tuning and Efficient Quantization for Optimization of Diffusion Models
Diffusion models have emerged as a critical component in the domain of deep generative models for image synthesis. However, the substantial computational demands of their practical deployment limit the viability of developing next-generation machine learning algorithms. This paper presents an optimization strategy that includes quantization, fine-tuning, and inference techniques to improve the effectiveness of diffusion models. It also discusses identifying and reducing particular training bottlenecks. By means of thorough testing and assessment, the suggested enhancements are methodically compared with baseline models. While preserving similar image quality, the optimization methods show improved computational efficiency. This advancement makes it easier to create diffusion models that are more precise and scalable, enabling wider uses in computer vision and related domains.
- Top Cited PapersOpen Access
- 18 August 2024
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 57 (10) , 1-51
Large language models (LLMs): survey, technical frameworks, and future challenges
- 18 August 2024
- Published by Springer Nature in Artificial Intelligence Review
- Vol. 57 (10) , 1-51
Top Cited PapersOpen AccessLarge language models (LLMs): survey, technical frameworks, and future challenges
- 25 June 2024
- Published by Springer Nature in International Journal of Multimedia Information Retrieval
- Vol. 13 (3) , 1-28
Adversarial attacks and defenses for large language models (LLMs): methods, frameworks & challenges
- 25 June 2024
- Published by Springer Nature in International Journal of Multimedia Information Retrieval
- Vol. 13 (3) , 1-28
Adversarial attacks and defenses for large language models (LLMs): methods, frameworks & challenges