arXiv subjects

Suraj Singh

Publications and source records attributed to Suraj Singh.

8 recordsLinked to original sources

arXiv2026

Worldline-Susceptibility Scheduling for Quantum Annealing Beyond Local-Adiabatic Evolution

The performance of quantum annealing depends critically on how the available annealing time is distributed along the evolution. Although the Roland Cerf local adiabatic schedule is theoretically optimal, it requires complete knowledge of the instantaneous spectral gap, making it impractical for large optimization problems. We propose a computationally inexpensive surrogate schedule based on the worldline magnetization susceptibility measured during simulated quantum annealing. The susceptibility is obtained directly from equilibrium Monte Carlo sampling and identifies the critical region of the anneal without requiring spectral information. Using exact diagonalization of Sherrington Kirkpatrick spin glass instances as ground truth, we show that the resulting schedule consistently outperforms conventional linear annealing and, for a substantial fraction of instances, also surpasses the exact Roland Cerf schedule. We demonstrate that this unexpected behaviour originates from two finite time failure modes of exact local adiabatic scheduling a boundary gap trap, in which the minimum spectral gap occurs at the end of the anneal, and an oscillatory instability caused by excessively localized time allocation around an interior minimum gap. These results suggest that robust scheduling based on inexpensive equilibrium observables can outperform exact spectral gap based strategies under realistic finite time conditions. The complete methodology is implemented in the open source Qanneal framework.

quant-ph↗

arXiv2026

FiMI: A Domain-Specific Language Model for Indian Finance Ecosystem

We present FiMI (Finance Model for India), a domain-specialized financial language model developed by National Payments Corporation of India (NPCI) for Indian digital payment systems. We develop two model variants: FiMI Base and FiMI Instruct. FiMI adapts the Mistral Small 24B architecture through a multi-stage training pipeline, beginning with continuous pre-training on 68 Billion tokens of curated financial, multilingual (English, Hindi, Hinglish), and synthetic data. This is followed by instruction fine-tuning and domain-specific supervised fine-tuning focused on multi-turn, tool-driven conversations that model real-world workflows, such as transaction disputes and mandate lifecycle management. Evaluations reveal that FiMI Base achieves a 20\% improvement over the Mistral Small 24B Base model on finance reasoning benchmark, while FiMI Instruct outperforms the Mistral Small 24B Instruct model by 87\% on domain-specific tool-calling. Moreover, FiMI achieves these significant domain gains while maintaining comparable performance to models of similar size on general benchmarks.

cs.AI↗

arXiv2025

Boundary-Aware Vision Transformer for Angiography Vascular Network Segmentation

Accurate segmentation of vascular structures in coronary angiography remains a core challenge in medical image analysis due to the complexity of elongated, thin, and low-contrast vessels. Classical convolutional neural networks (CNNs) often fail to preserve topological continuity, while recent Vision Transformer (ViT)-based models, although strong in global context modeling, lack precise boundary awareness. In this work, we introduce BAVT, a Boundary-Aware Vision Transformer, a ViT-based architecture enhanced with an edge-aware loss that explicitly guides the segmentation toward fine-grained vascular boundaries. Unlike hybrid transformer-CNN models, BAVT retains a minimal, scalable structure that is fully compatible with large-scale vision foundation model (VFM) pretraining. We validate our approach on the DCA-1 coronary angiography dataset, where BAVT achieves superior performance across medical image segmentation metrics outperforming both CNN and hybrid baselines. These results demonstrate the effectiveness of combining plain ViT encoders with boundary-aware supervision for clinical-grade vascular segmentation.

cs.CV↗

arXiv2025

DIPLI: Deep Image Prior Lucky Imaging for Blind Astronomical Image Restoration

Modern image restoration and super-resolution methods utilize deep learning due to its superior performance compared to traditional algorithms. However, deep learning typically requires large labeled training datasets, which are rarely available in astrophotography. Deep Image Prior (DIP) bypasses this constraint by performing unsupervised optimization on a single image without training data; however, DIP often suffers from overfitting, artifact generation, and instability. This work proposes DIPLI - a framework designed specifically for resolved, high-contrast astronomical targets that shifts from single-frame to multi-frame processing using the Back Projection technique, combined with dense optical flow estimation via the TVNet model, and replaces deterministic predictions with Monte Carlo estimation obtained through Stochastic Gradient Langevin Dynamics (SGLD). A comprehensive evaluation compares the method against the original DIP, the transformer-based model RVRT, and the diffusion-based model DiffIR2VR-Zero on synthetic data with ground truth, while comparing qualitatively against Lucky Imaging on real astronomical data. On synthetic datasets, DIPLI achieves the best perceptual fidelity scores (LPIPS in 12/12 and DISTS in 10/12 scenarios), while the diffusion-based DiffIR2VR-Zero achieves the best pixel-level distortion scores (PSNR in 9/12 and SSIM in 8/12 scenarios), consistent with the well-known perceptual-distortion trade-off in image restoration. Compared to classical Lucky Imaging, the model requires far fewer input frames (7-13 versus thousands) and avoids the need for early stopping that limits standard DIP. Qualitative evaluation on real-world data of resolved solar-system objects, where ground truth is unavailable and domain shifts typically hinder generalization, suggests that the method appears to preserve fine detail while suppressing noise and artifacts.

cs.CV↗

arXiv2024

SOAR: Advancements in Small Body Object Detection for Aerial Imagery Using State Space Models and Programmable Gradients

Small object detection in aerial imagery presents significant challenges in computer vision due to the minimal data inherent in small-sized objects and their propensity to be obscured by larger objects and background noise. Traditional methods using transformer-based models often face limitations stemming from the lack of specialized databases, which adversely affect their performance with objects of varying orientations and scales. This underscores the need for more adaptable, lightweight models. In response, this paper introduces two innovative approaches that significantly enhance detection and segmentation capabilities for small aerial objects. Firstly, we explore the use of the SAHI framework on the newly introduced lightweight YOLO v9 architecture, which utilizes Programmable Gradient Information (PGI) to reduce the substantial information loss typically encountered in sequential feature extraction processes. The paper employs the Vision Mamba model, which incorporates position embeddings to facilitate precise location-aware visual understanding, combined with a novel bidirectional State Space Model (SSM) for effective visual context modeling. This State Space Model adeptly harnesses the linear complexity of CNNs and the global receptive field of Transformers, making it particularly effective in remote sensing image classification. Our experimental results demonstrate substantial improvements in detection accuracy and processing efficiency, validating the applicability of these approaches for real-time small object detection across diverse aerial scenarios. This paper also discusses how these methodologies could serve as foundational models for future advancements in aerial object recognition technologies. The source code will be made accessible here.

cs.CV↗

arXiv2023

ICML 2023 Topological Deep Learning Challenge : Design and Results

This paper presents the computational challenge on topological deep learning that was hosted within the ICML 2023 Workshop on Topology and Geometry in Machine Learning. The competition asked participants to provide open-source implementations of topological neural networks from the literature by contributing to the python packages TopoNetX (data processing) and TopoModelX (deep learning). The challenge attracted twenty-eight qualifying submissions in its two-month duration. This paper describes the design of the challenge and summarizes its main findings.

cs.LG↗

arXiv2021

Magnetodynamic properties of dipole-coupled 1D magnonic crystals

Magnonic crystals are magnetic metamaterials, that provide a promising way to manipulate magnetodynamic properties by controlling the geometry of the patterned structures. Here, we study the magnetodynamic properties of 1D magnonic crystals consisting of parallel NiFe strips with different strip widths and separations. The strips couple via dipole-dipole interactions. As an alternative to experiments and/or micromagnetic simulations, we investigate the accuracy of a simple macrospin model. For the case of simple strips, a model with a single free parameter to account for an overestimation of the out-of-plane demagnetization of the magnonic lattice is described. By adjusting this parameter a good fit with experimental as well as micromagnetic results is obtained. Moreover, the Gilbert damping is found independent of the lattice constant however the inhomogeneous linewidth broadening found to increase with decreasing stripe separation.

cond-mat.mes-hall↗

arXiv2020

Field induced single molecule magnet behavior in Dy-based coordination polymer

A new mononuclear Dysprosium based Coordination Polymer {Dy-CP} was investigated magnetically using dc and ac magnetic susceptibility. The dc magnetic susceptibility does not exhibit any long-range ordering down to 1.8 K and the negative value of Curie Constant (~ - 4 K) indicate the dominance of antiferromagnetic interactions between the Dy (III) spins. Ac susceptibility exhibits absence of single molecular magnet behavior at zero dc magnetic field and shows signal of quantum tunneling magnetization (QTM) below 8 K. However, on the superimposition of dc magnetic field (3 kOe), frequency dependent relaxation peak emerged at T_f = 5 K and QTM signal suppress at higher fields. The intermediate value of Mydosh parameter calculated from the shift in peak position (T_f) in ac susceptibility reflects the formation of superparamagnetic state. Further, the temperature dependence of Tf is analyzed with Arrhenius and Cole-Cole plot. The magnetic susceptibility analysis yields characteristics pre-relaxation factor 1.40x10^(-12) sec and energy barrier ΔE/k_B = 93.4 K, indicating the slow spin relaxation. The Cole-Cole fit to the ac susceptibility data shows further evidence for the single ion spin relaxation. Thus, the magnetic measurements support the single-molecule magnet behavior in Dy-CP under the application of dc magnetic field.

cond-mat.mtrl-sci↗