arXiv subjects

Amar Kumar

Publications and source records attributed to Amar Kumar.

17 recordsLinked to original sources

arXiv2025

Discovering Latent Graphs with GFlowNets for Diverse Conditional Image Generation

Capturing diversity is crucial in conditional and prompt-based image generation, particularly when conditions contain uncertainty that can lead to multiple plausible outputs. To generate diverse images reflecting this diversity, traditional methods often modify random seeds, making it difficult to discern meaningful differences between samples, or diversify the input prompt, which is limited in verbally interpretable diversity. We propose Rainbow, a novel conditional image generation framework, applicable to any pretrained conditional generative model, that addresses inherent condition/prompt uncertainty and generates diverse plausible images. Rainbow is based on a simple yet effective idea: decomposing the input condition into diverse latent representations, each capturing an aspect of the uncertainty and generating a distinct image. First, we integrate a latent graph, parameterized by Generative Flow Networks (GFlowNets), into the prompt representation computation. Second, leveraging GFlowNets' advanced graph sampling capabilities to capture uncertainty and output diverse trajectories over the graph, we produce multiple trajectories that collectively represent the input condition, leading to diverse condition representations and corresponding output images. Evaluations on natural image and medical image datasets demonstrate Rainbow's improvement in both diversity and fidelity across image synthesis, image generation, and counterfactual generation tasks.

cs.CV↗

arXiv2025

Ab-initio investigation of the interfacial structural, electronic, and magnetic properties of Co$_{2}$MnAl/X (X = MgO and GaAs) heterostructures

The structural, electronic, and magnetic properties of (100)-oriented Co$_{2}$MnAl/MgO and Co$_{2}$MnAl/GaAs heterostructures are investigated using plane-wave pseudopotential density functional theory. For the Co$_{2}$MnAl/MgO, CoCo-MgMg, CoCo-OO, MnAl-MgMg, and MnAl-OO interfaces in top-to-top configurations are studied, while for Co$_{2}$MnAl/GaAs, both top-to-top (Co-Ga, Co-As, Mn-Ga, Mn-As, Al-Ga, Al-As) and bridge-site (CoCo-Ga, CoCo-As, MnAl-Ga, MnAl-As) interfaces are considered. The interfacial geometries featuring Co- or CoCo-atomic terminations for the Co2MnAl slab exhibit larger adhesion energies compared to those terminated with Mn-, Al-, or MnAl-atomic terminations. This indicates their greater interfacial stability. In contrast, MnAl-, Mn-, or Al-terminated interfaces preserve near half-metallicity, whereas Co- and CoCo-terminated geometries display a strongly metallic character. All studied interfaces show enhanced magnetic moments relative to their bulk counterparts, primarily arising from interfacial atoms and their nearest neighbours. These findings offer valuable insights for optimizing Co2MnAl-based heterostructures in spintronic applications.

cond-mat.mtrl-sci↗

arXiv2025

Pixels Under Pressure: Exploring Fine-Tuning Paradigms for Foundation Models in High-Resolution Medical Imaging

Advancements in diffusion-based foundation models have improved text-to-image generation, yet most efforts have been limited to low-resolution settings. As high-resolution image synthesis becomes increasingly essential for various applications, particularly in medical imaging domains, fine-tuning emerges as a crucial mechanism for adapting these powerful pre-trained models to task-specific requirements and data distributions. In this work, we present a systematic study, examining the impact of various fine-tuning techniques on image generation quality when scaling to high resolution 512x512 pixels. We benchmark a diverse set of fine-tuning methods, including full fine-tuning strategies and parameter-efficient fine-tuning (PEFT). We dissect how different fine-tuning methods influence key quality metrics, including Fr\'echet Inception Distance (FID), Vendi score, and prompt-image alignment. We also evaluate the utility of generated images in a downstream classification task under data-scarce conditions, demonstrating that specific fine-tuning strategies improve both generation fidelity and downstream performance when synthetic images are used for classifier training and evaluation on real images. Our code is accessible through the project website - https://tehraninasab.github.io/PixelUPressure/.

eess.IV↗

arXiv2025

AURA: A Multi-Modal Medical Agent for Understanding, Reasoning & Annotation

Recent advancements in Large Language Models (LLMs) have catalyzed a paradigm shift from static prediction systems to agentic AI agents capable of reasoning, interacting with tools, and adapting to complex tasks. While LLM-based agentic systems have shown promise across many domains, their application to medical imaging remains in its infancy. In this work, we introduce AURA, the first visual linguistic explainability agent designed specifically for comprehensive analysis, explanation, and evaluation of medical images. By enabling dynamic interactions, contextual explanations, and hypothesis testing, AURA represents a significant advancement toward more transparent, adaptable, and clinically aligned AI systems. We highlight the promise of agentic AI in transforming medical image analysis from static predictions to interactive decision support. Leveraging Qwen-32B, an LLM-based architecture, AURA integrates a modular toolbox comprising: (i) a segmentation suite with phase grounding, pathology segmentation, and anatomy segmentation to localize clinically meaningful regions; (ii) a counterfactual image-generation module that supports reasoning through image-level explanations; and (iii) a set of evaluation tools including pixel-wise difference-map analysis, classification, and advanced state-of-the-art components to assess diagnostic relevance and visual interpretability.

cs.CV↗

arXiv2025

Pixel Perfect MegaMed: A Megapixel-Scale Vision-Language Foundation Model for Generating High Resolution Medical Images

Medical image synthesis presents unique challenges due to the inherent complexity and high-resolution details required in clinical contexts. Traditional generative architectures such as Generative Adversarial Networks (GANs) or Variational Auto Encoder (VAEs) have shown great promise for high-resolution image generation but struggle with preserving fine-grained details that are key for accurate diagnosis. To address this issue, we introduce Pixel Perfect MegaMed, the first vision-language foundation model to synthesize images at resolutions of 1024x1024. Our method deploys a multi-scale transformer architecture designed specifically for ultra-high resolution medical image generation, enabling the preservation of both global anatomical context and local image-level details. By leveraging vision-language alignment techniques tailored to medical terminology and imaging modalities, Pixel Perfect MegaMed bridges the gap between textual descriptions and visual representations at unprecedented resolution levels. We apply our model to the CheXpert dataset and demonstrate its ability to generate clinically faithful chest X-rays from text prompts. Beyond visual quality, these high-resolution synthetic images prove valuable for downstream tasks such as classification, showing measurable performance gains when used for data augmentation, particularly in low-data regimes. Our code is accessible through the project website - https://tehraninasab.github.io/pixelperfect-megamed.

eess.IV↗

arXiv2025

Leveraging Vision-Language Foundation Models to Reveal Hidden Image-Attribute Relationships in Medical Imaging

Vision-language foundation models (VLMs) have shown impressive performance in guiding image generation through text, with emerging applications in medical imaging. In this work, we are the first to investigate the question: 'Can fine-tuned foundation models help identify critical, and possibly unknown, data properties?' By evaluating our proposed method on a chest x-ray dataset, we show that these models can generate high-resolution, precisely edited images compared to methods that rely on Structural Causal Models (SCMs) according to numerous metrics. For the first time, we demonstrate that fine-tuned VLMs can reveal hidden data relationships that were previously obscured due to available metadata granularity and model capacity limitations. Our experiments demonstrate both the potential of these models to reveal underlying dataset properties while also exposing the limitations of fine-tuned VLMs for accurate image editing and susceptibility to biases and spurious correlations.

cs.CV↗

arXiv2025

Language-Guided Trajectory Traversal in Disentangled Stable Diffusion Latent Space for Factorized Medical Image Generation

Text-to-image diffusion models have demonstrated a remarkable ability to generate photorealistic images from natural language prompts. These high-resolution, language-guided synthesized images are essential for the explainability of disease or exploring causal relationships. However, their potential for disentangling and controlling latent factors of variation in specialized domains like medical imaging remains under-explored. In this work, we present the first investigation of the power of pre-trained vision-language foundation models, once fine-tuned on medical image datasets, to perform latent disentanglement for factorized medical image generation and interpolation. Through extensive experiments on chest X-ray and skin datasets, we illustrate that fine-tuned, language-guided Stable Diffusion inherently learns to factorize key attributes for image generation, such as the patient's anatomical structures or disease diagnostic features. We devise a framework to identify, isolate, and manipulate key attributes through latent space trajectory traversal of generative models, facilitating precise control over medical image synthesis.

cs.CV↗

arXiv2025

RL4Med-DDPO: Reinforcement Learning for Controlled Guidance Towards Diverse Medical Image Generation using Vision-Language Foundation Models

Vision-Language Foundation Models (VLFM) have shown a tremendous increase in performance in terms of generating high-resolution, photorealistic natural images. While VLFMs show a rich understanding of semantic content across modalities, they often struggle with fine-grained alignment tasks that require precise correspondence between image regions and textual descriptions, a limitation in medical imaging, where accurate localization and detection of clinical features are essential for diagnosis and analysis. To address this issue, we propose a multi-stage architecture where a pre-trained VLFM (e.g. Stable Diffusion) provides a cursory semantic understanding, while a reinforcement learning (RL) algorithm refines the alignment through an iterative process that optimizes for understanding semantic context. The reward signal is designed to align the semantic information of the text with synthesized images. Experiments on the public ISIC2019 skin lesion dataset demonstrate that the proposed method improves (a) the quality of the generated images, and (b) the alignment with the text prompt over the original fine-tuned Stable Diffusion baseline. We also show that the synthesized samples could be used to improve disease classifier performance for underrepresented subgroups through augmentation. Our code is accessible through the project website: https://parhamsaremi.github.io/rl4med-ddpo

cs.CV↗

arXiv2025

Surface properties of Co$_2$MnAl Heusler alloy

Using the plane-wave pseudopotential method within the framework of density functional theory, Co$_2$MnAl (100), (110), and (111) surfaces with different atomic terminations have been studied in the context of some key spintronics properties, viz., surface energy, half-metallicity, magnetization, and magnetic anisotropy. The present study reveals that the MnAl-(100), Co-Al-(111), and Al-(111) surfaces exhibit negative surface energies over a wide range of chemical potentials, indicating their strong structural stability. The MnAl-(100), CoCoMnAl-(110), and Co-Mn-(111) surfaces maintain the nearly half-metallic nature like the bulk-Co$_2$MnAl, while this nearly half-metallic nature even improved for the Al-(111) surface. In contrast, the rest of the considered surfaces, CoCo-(100), Co-Al-(111) and Mn-(111) surfaces, display the strong metallic nature. Magnetization is enhanced for most surface configurations, except for Al-(111), where it decreases due to reduced moments of the exterior atoms. Regarding magnetic anisotropy, only the MnAl-(100) and Co-Mn-(111) surfaces exhibit the positive magneto-crystalline anisotropy of $\sim$0.23 and $\sim$0.33 mJ/m2, respectively. All these findings suggest that the Co-Mn-(111) and MnAl-(100) surfaces are quite appealing for spintronics applications, considering the structural stability, electronic properties, and magnetic anisotropy.

cond-mat.mtrl-sci↗

arXiv2025

PRISM: High-Resolution & Precise Counterfactual Medical Image Generation using Language-guided Stable Diffusion

Developing reliable and generalizable deep learning systems for medical imaging faces significant obstacles due to spurious correlations, data imbalances, and limited text annotations in datasets. Addressing these challenges requires architectures that are robust to the unique complexities posed by medical imaging data. Rapid advancements in vision-language foundation models within the natural image domain prompt the question of how they can be adapted for medical imaging tasks. In this work, we present PRISM, a framework that leverages foundation models to generate high-resolution, language-guided medical image counterfactuals using Stable Diffusion. Our approach demonstrates unprecedented precision in selectively modifying spurious correlations (the medical devices) and disease features, enabling the removal and addition of specific attributes while preserving other image characteristics. Through extensive evaluation, we show how PRISM advances counterfactual generation and enables the development of more robust downstream classifiers for clinically deployable solutions. To facilitate broader adoption and research, we make our code publicly available at https://github.com/Amarkr1/PRISM.

cs.CV↗

arXiv2024

Effect of the Lattice-distortion on the Electronic Structure, Magnetic Anisotropy, and Hall Conductivities of the CoFeCrGa Spin Gapless Semiconductor: A First-Principles Study

Spin gapless semiconductors (SGSs), novel quantum materials, are notable for their tunable spin-transport properties. Considering that the SGS materials might have an invariably deformed lattice upon integration into devices, and given that the SGS nature is highly sensitive to external factors, the impact of lattice distortions on the different physical properties of CoFeCrGa SGS alloy has been investigated using density functional theory calculations. For lattice distortions, the uniform strain corresponding to $-6\% \leq \Delta V / V_0 \leq 6\% \quad (a: 5.60\text{-}5.83~\text\r{A})$, and the tetragonal distortion corresponding to $0.8 \leq c/a \leq 1.2 \quad (a: 5.38\text{-}6.16~\text\r{A},~c: 4.92\text{-}6.45~\text\r{A})$ are modelled. All uniformly strained CoFeCrGa structures are found to display SGS character, magnetic isotropy, small anomalous Hall conductivity (AHC), and small spin Hall conductivity (SHC) - closely resembling those of the ideal CoFeCrGa structure. In contrast, the tetragonally deformed structures display nearly half-metallic behavior with very high spin polarization, very large magnetic anisotropy ($ \sim 10^6~\mathrm{J/m^3}$), and very large AHC ranging from ($ -215 \text{ to } 250~\mathrm{S/cm} $) depending on the axial ratio of the distorted structure. The SHC, however, does not change significantly under tetragonal distortion and remains nearly of the same order as that of the Y-I ordered structure. In summary, these findings demonstrate that CoFeCrGa displays favorable spintronic properties even under lattice distortions, underscoring its potential for next-generation spintronic applications.

cond-mat.mtrl-sci↗

arXiv2024

DeCoDEx: Confounder Detector Guidance for Improved Diffusion-based Counterfactual Explanations

Deep learning classifiers are prone to latching onto dominant confounders present in a dataset rather than on the causal markers associated with the target class, leading to poor generalization and biased predictions. Although explainability via counterfactual image generation has been successful at exposing the problem, bias mitigation strategies that permit accurate explainability in the presence of dominant and diverse artifacts remain unsolved. In this work, we propose the DeCoDEx framework and show how an external, pre-trained binary artifact detector can be leveraged during inference to guide a diffusion-based counterfactual image generator towards accurate explainability. Experiments on the CheXpert dataset, using both synthetic artifacts and real visual artifacts (support devices), show that the proposed method successfully synthesizes the counterfactual images that change the causal pathology markers associated with Pleural Effusion while preserving or ignoring the visual artifacts. Augmentation of ERM and Group-DRO classifiers with the DeCoDEx generated images substantially improves the results across underrepresented groups that are out of distribution for each class. The code is made publicly available at https://github.com/NimaFathi/DeCoDEx.

cs.CV↗

arXiv2023

Effect of Point Defects and Lattice Distortions on the Structural, Electronic, and Magnetic properties of Co$_2$MnAl Heusler alloy

The effects of various point defects and lattice distortions on the structural, electronic, and magnetic properties of Co$_2$MnAl alloy are investigated using density functional theory calculations. For the point defects, six types of binary antisites, three types of ternary antisites, and three kinds of vacancies have been simulated with different disorder degrees, up to a maximum of 12.50%. For the lattice distortions, cubic strain within -10% $\leq$$\Delta{V/V_0}$$\leq$ 10% (corresponding to 5.50\r{A} $\leq$ a $\leq$5.88\r{A}) and tetragonal distortions with 0.5$\leq$$\textit{c/a}$$\leq$1.5 at three different unit-cell volumes - $\textit{V}_0$ and ($\textit{V}_0$$\pm5$%$\textit{V}_0$) have been considered. The Co$_{Al}$ and Mn$_{Al}$ binary antisite disordered structures (namely, Co$_{2.0625}$MnAl$_{0.9375}$, Co$_{2.125}$MnAl$_{0.875}$, Co$_2$Mn$_{1.0625}$Al$_{0.9375}$ and Co$_2$Mn$_{1.125}$Al$_{0.875}$) and (Co$_{Al}$+Mn$_{Al}$) ternary antisite disordered structure (Co$_{2.0625}$Mn$_{1.0625}$Al$_{0.875}$) exhibit perfect half-metallicity. The rest of the antisite disorders have a marginal effect on the half-metallic properties of Co$_2$MnAl, along with high spin polarization ($\textit{P}$ $\geq$ 70%) and nearly same magnetization ($\textit{M$_s$}$) as that for ideal structure. Conversely, the vacancy defects significantly affect the electronic and magnetic properties. The cubic strained structures exhibit high $\textit{P}$ and constant $\textit{M$_s$}$. Under negative strain within -10% $\leq$$\Delta{V/V_0}$$\leq$ -7% (for 5.50\r{A} $\leq$ a $\leq$ 5.58\r{A}), the strained structures have perfect half-metallicity. On the other hand, tetragonal distortions lead to significant degradation in half-metallic behavior, except for small distortion values $\Delta{c/a}$, irrespective of their volume.

cond-mat.mtrl-sci↗

arXiv2023

Debiasing Counterfactuals In the Presence of Spurious Correlations

Deep learning models can perform well in complex medical imaging classification tasks, even when basing their conclusions on spurious correlations (i.e. confounders), should they be prevalent in the training dataset, rather than on the causal image markers of interest. This would thereby limit their ability to generalize across the population. Explainability based on counterfactual image generation can be used to expose the confounders but does not provide a strategy to mitigate the bias. In this work, we introduce the first end-to-end training framework that integrates both (i) popular debiasing classifiers (e.g. distributionally robust optimization (DRO)) to avoid latching onto the spurious correlations and (ii) counterfactual image generation to unveil generalizable imaging markers of relevance to the task. Additionally, we propose a novel metric, Spurious Correlation Latching Score (SCLS), to quantify the extent of the classifier reliance on the spurious correlation as exposed by the counterfactual images. Through comprehensive experiments on two public datasets (with the simulated and real visual artifacts), we demonstrate that the debiasing method: (i) learns generalizable markers across the population, and (ii) successfully ignores spurious correlations and focuses on the underlying disease pathology.

cs.CV↗

arXiv2023

Photochargeable Li-ion Battery: Type II Heterojunction Exposes Underlying Band Gap in 'Metallic' Lithiated MoS2

Light chargeable metal-ion batteries using semiconductor heterostructures are gaining enormous interest. A few such heterostructures such as MoS2/MoOy and TiS2/TiO2 have been shown to function as photocathodes in photochargeable Li-ion batteries, where the type II set-up has been proposed to generate spatially separated (longer-lived) excitons upon photo-exposure. The Li intercalated MoS2, generated during the discharge cycle of the battery, undergoes a phase transition from the semiconducting (2H) to a metallic (1T) phase, in contrast to its TiS2 counterpart, casting a doubt over the photocharging process. Here, employing density functional theory based traditional as well as unconventional computational schemes along with relevant spectroscopic techniques, we show that in LixMoS2/MoO3 heterostructure an underlying band gap of LixMoS2 is exposed, upto a certain value of x, due to dispersion of electron density onto MoO3 justifying the observed photocharging. We believe that the general concepts explored in this study will be important in the rational design of photo-cathode materials in Li-ion batteries.

cond-mat.mtrl-sci↗

arXiv2023

Photo-Rechargeable Li Ion Batteries using TiS2 Cathode

Photo-rechargeable (solar) battery can be considered as an energy harvesting cum storage system, where it can charge the conventional metal-ion battery using light instead of electricity, without having other parasitic reactions. Here we demonstrate a two-electrode lithium ion solar battery with multifaceted TiS2-TiO2 hybrid sheets as cathode. Choice of TiS2-TiO2 electrode ensures the formation of a type II semiconductor heterostructure while the lateral heterostructure geometry ensures high mass/charge transfer and light interactions with the electrode. TiS2 has a higher lithium binding energy (1.6 eV) than TiO2 (1.03 eV), ensuring the possibilities of higher amount of Li ion insertion to TiS2 and hence the maximum recovery with the photocharging, as further confirmed by the experiments. Apart from the demonstration of solar solid-state batteries, the charging of lithium ion full cell with light indicates the formation of lithium intercalated graphite compounds, ensuring the charging of the battery without any other parasitic reactions at the electrolyte or electrode-electrolyte interfaces. Possible mechanisms proposed here for the charging and discharging processes of solar batteries, based on our experimental and theoretical results, indicate the potential of such systems in forthcoming era of renewable energies.

physics.app-ph↗

arXiv2022

Counterfactual Image Synthesis for Discovery of Personalized Predictive Image Markers

The discovery of patient-specific imaging markers that are predictive of future disease outcomes can help us better understand individual-level heterogeneity of disease evolution. In fact, deep learning models that can provide data-driven personalized markers are much more likely to be adopted in medical practice. In this work, we demonstrate that data-driven biomarker discovery can be achieved through a counterfactual synthesis process. We show how a deep conditional generative model can be used to perturb local imaging features in baseline images that are pertinent to subject-specific future disease evolution and result in a counterfactual image that is expected to have a different future outcome. Candidate biomarkers, therefore, result from examining the set of features that are perturbed in this process. Through several experiments on a large-scale, multi-scanner, multi-center multiple sclerosis (MS) clinical trial magnetic resonance imaging (MRI) dataset of relapsing-remitting (RRMS) patients, we demonstrate that our model produces counterfactuals with changes in imaging features that reflect established clinical markers predictive of future MRI lesional activity at the population level. Additional qualitative results illustrate that our model has the potential to discover novel and subject-specific predictive markers of future activity.

cs.CV↗