arXiv subjects

Priyanka Singh

Publications and source records attributed to Priyanka Singh.

At least 19 recordsLinked to original sources

arXiv2026

Probing the Baryon Distribution with Fast Radio Bursts

Baryonic feedback redistributes matter on small to mid cosmological scales, ultimately limiting inferences from Stage IV galaxy surveys. Direct baryon tracers are crucial for recovering cosmological signals masked by astrophysical effects, and vice versa: galaxy formation and other astrophysical processes must be interpreted cosmologically. Fast radio bursts (FRBs) serve as such tracers: their dispersion measure (DM) records the line-of-sight integrated ionised electron density. The Square Kilometre Array (SKA) will be the only radio telescope capable of detecting many FRBs in the southern hemisphere, significantly enhancing synergy with surveys such as Rubin Observatory. This chapter completes the FRB trilogy by forecasting the SKA's potential to constrain the baryon distribution from cosmological to galactic scales and across cosmic time. We tackle this question by investigating the DM scatter as a function of redshift. We also study the statistical properties of the DM field and its cross-correlation with Stage IV galaxy surveys. Our focus is on cosmic shear and galaxy clustering. This shows that the SKA can play a crucial role in pinpointing baryonic feedback models, thereby greatly enhancing the cosmological constraining power of Stage IV galaxy surveys. Furthermore, we show that the SKA will be able to measure the properties of the circumgalactic medium using the scattering timescale of FRBs. Lastly, the large redshift range of FRB detections with the SKA can improve our understanding of the epoch of reionisation. It may also clarify the mechanism behind FRBs.

astro-ph.CO↗

arXiv2026

Extending Detection Engineering to Digital Forensics: The Velociraptor Unified Detection-Forensics Methodology

Detection engineering and digital forensics have evolved in parallel rather than in partnership, leaving a gap between real-time alerting and forensic analysis. This paper develops a unified detection-forensics methodology using Velociraptor, where detection logic directly initiates targeted evidence acquisition at the point of detection. The contribution is threefold: (1) a four-stage methodology (baseline establishment, evidence correlation, attack chain analysis, and scenario labelling with confidence) that converts artefact knowledge into reusable and testable detection rules suitable for both post-incident triage and live monitoring; (2) a practical demonstration, using three Velociraptor BaseVQL log sources (/forensics/windows/prefetch, /forensics/windows/usn, and /windows/wmi) that practitioners can deploy today, showing that artefact-based detections enable scalable forensic triage without full disk acquisition; and (3) evidence that periodic artefact analysis offers continuous monitoring while substantially reducing data volume compared to conventional endpoint logging. Two case studies illustrate the approach: a Prefetch/USN baseline for triage when Windows Event Logs are cleared or unavailable, and a WMI persistence correlation supporting both triage and continuous monitoring through periodic artefact analysis.

cs.CR↗

arXiv2026

Fast Radio Bursts as Cosmological Probes

Fast radio bursts (FRBs) are brief, coherent radio pulses of extragalactic origin. They typically last from microseconds to milliseconds and have energies large enough to be visible over cosmological distances. Since FRBs interact with free electrons along their paths, the original burst is dispersed (Dispersion Measure, DM) and broadened (scattering). Furthermore, the burst's polarization is altered by Faraday rotation. Consequently, FRBs are excellent probes of the cosmological distribution of baryons, the expansion of the Universe, magnetic fields, and minuscule effects of fundamental physics that accumulate over vast distances. This chapter is the second of a trilogy of FRB chapters and discusses FRBs as a standalone probe. We first introduce the foundation of FRB observables related to those questions. Next, we lay the groundwork for forecasting SKA's potential by describing the method to simulate the expected FRB population observable with the SKA. These synthetic FRB catalogues are then used to investigate the SKA's potential to probe the Universe's expansion rate and fundamental physics, such as the equivalence principle and the existence of massive photons. Furthermore, we investigate the possibility of tracing cosmic magnetic fields and investigating different dark matter candidates.

astro-ph.CO↗

arXiv2026

T-IMPACT: A Severity-Aware Benchmark for Contextual Image-Text Manipulation

Recent advances in vision-language models and generative editing systems have made it increasingly easy to produce persuasive multimodal misinformation by altering images, text, or both jointly. However, existing datasets focus mainly on authenticity, out-of-context mismatch, or manipulation type, and rarely capture how strongly an edit changes the likely interpretation of a post. We introduce T-IMPACT, a first-release severity-aware benchmark for manipulated news-style image-text pairs. T-IMPACT contains 98,786 examples spanning pristine, image-only, text-only, and joint manipulations, with a calibrated continuous severity signal, coarse low/medium/high labels, and supporting grounding metadata. Starting from a news image-text pair, the pipeline extracts semantic anchors, grounds them spatially, performs localized image edits and constrained caption rewrites, and calibrates contextual-impact scores using limited human ratings. In this release, the calibrated continuous score is the primary severity target, while the low/medium/high bands should be interpreted as coarse operating buckets rather than balanced classes. Experiments show that current models recover some authenticity signal, but severity prediction remains substantially harder and only weakly aligned with human judgment. T-IMPACT provides an initial benchmark for studying multimodal manipulation beyond binary real/fake classification toward graded contextual impact.

cs.CV↗

arXiv2026

Multi-Adapter Representation Interventions via Energy Calibration

Representation intervention has emerged as a promising paradigm for aligning large language models toward desired behaviors without modifying model weights. Existing methods typically apply a fixed intervention uniformly across all inputs. However, we find that the appropriate intervention direction and strength vary substantially across samples, and such indiscriminate intervention leads to degradation of general capabilities on benign inputs. To address these challenges, we propose Multi-Adapter Representation Interventions via Energy Calibration (MARI). Specifically, we introduce a competitive multi-adapter mechanism in which specialized experts capture non-linear correction patterns and adaptively determine the appropriate intervention direction and strength for different samples. Furthermore, we design an energy-based gating module that leverages internal propagation dynamics to distinguish inputs that are applicable for intervention. Extensive experiments across diverse model families and parameter scales demonstrate that MARI achieves state-of-the-art alignment performance. Our method significantly improves performance on TruthfulQA, BBQ, and safety benchmarks, while maintaining and even improving general capabilities on tasks such as MMLU and ARC. Our code is available at https://github.com/V1centNevwake/MARI.

cs.AI↗

arXiv2026

A Fixed-Time Sliding-Mode Framework for Constraint Optimization

This paper develops a robust fixed time optimization framework for constrained problems that guarantees exact constraint satisfaction and convergence to KKT points within fixed time , independent of initial conditions. The approach treats the Lagrange multipliers as control inputs, composed of an equivalent control and a switching control, with the system states representing the decision variables. An equivalent control steers the gradient flow to a local KKT point asymptotically for nonconvex objectives and to unique global optimum in fixed time for convex objectives. Constraint enforcement is achieved by embedding the equality constraints directly as a sliding manifold, with a fixed time switching control ensuring rapid and reliable feasibility. The framework further accounts for the matched disturbances, providing robustness guarantees that are theoretically characterized and illustrated using spherical constraints. Numerical studies on a 3-bus AC optimal power flow problem and distributed consensus=based parameter estimation problem demonstrate the effectiveness, scalability and robustness of proposed approach.

math.OC↗

arXiv2026

Unsupervised Domain Adaptation for Audio Deepfake Detection with Modular Statistical Transformations

Audio deepfake detection systems trained on one dataset often fail when deployed on data from different sources due to distributional shifts in recording conditions, synthesis methods, and acoustic environments. We present a modular pipeline for unsupervised domain adaptation that combines pre-trained Wav2Vec 2.0 embeddings with statistical transformations to improve cross-domain generalization without requiring labeled target data. Our approach applies power transformation for feature normalization, ANOVA-based feature selection, joint PCA for domain-agnostic dimensionality reduction, and CORAL alignment to match source and target covariance structures before classification via logistic regression. We evaluate on two cross-domain transfer scenarios: ASVspoof 2019 LA to Fake-or-Real (FoR) and FoR to ASVspoof, achieving 62.7--63.6\% accuracy with balanced performance across real and fake classes. Systematic ablation experiments reveal that feature selection (+3.5%) and CORAL alignment (+3.2%) provide the largest individual contributions, with the complete pipeline improving accuracy by 10.7% over baseline. While performance is modest compared to within-domain detection (94-96%), our pipeline offers transparency and modularity, making it suitable for deployment scenarios requiring interpretable decisions.

cs.SD↗

arXiv2026

D-SECURE: Dual-Source Evidence Combination for Unified Reasoning in Misinformation Detection

Multimodal misinformation increasingly mixes realistic im-age edits with fluent but misleading text, producing persuasive posts that are difficult to verify. Existing systems usually rely on a single evidence source. Content-based detectors identify local inconsistencies within an image and its caption but cannot determine global factual truth. Retrieval-based fact-checkers reason over external evidence but treat inputs as coarse claims and often miss subtle visual or textual manipulations. This separation creates failure cases where internally consistent fabrications bypass manipulation detectors and fact-checkers verify claims that contain pixel-level or token-level corruption. We present D-SECURE, a framework that combines internal manipulation detection with external evidence-based reasoning for news-style posts. D-SECURE integrates the HAMMER manipulation detector with the DEFAME retrieval pipeline. DEFAME performs broad verification, and HAMMER analyses residual or uncertain cases that may contain fine-grained edits. Experiments on DGM4 and ClaimReview samples highlight the complementary strengths of both systems and motivate their fusion. We provide a unified, explainable report that incorporates manipulation cues and external evidence.

cs.CV↗

arXiv2025

Impact of Galaxy Cluster Environment on the Stellar Mass Function of Galaxies

Galaxy clusters represent some of the most extreme environments in the Universe. They are ideal locations to study the impact of an extreme environment on the evolution of the Stellar Mass Function (SMF), which describes the statistical distribution of galaxies as a function of their stellar masses. In this work, we examine how the SMF of galaxies depends on factors such as the surrounding environments, whether they reside in isolated fields or clusters. We use the 9-band photometric galaxy data of the G9 patch from the Kilo Degree Survey (optical) and the VISTA Kilo-Degree Infrared Galaxy Survey (infrared), containing around 3.7 million galaxies, overlapping with the cluster catalog provided by the eROSITA Final Equatorial Depth Surveys (eFEDS). After applying appropriate selection criteria, we have 105 eFEDS clusters within the redshift range 0.385-0.8, covering $\sim 46$ square degrees. The large, continuous overlap of the surveys allows us to examine the SMF of the cluster galaxies within the cluster-centric radial bins up to $5R_{500}$. We find a clear detection of the cluster galaxy SMF up to $2R_{500}$ beyond which it's consistent with the background. We divide the cluster sample into redshift, mass, and X-ray luminosity bins to examine their impact on the SMF. The SMF of cluster galaxies for the high-mass clusters shows a decline at low stellar masses ($M_*\lesssim 2\times 10^{10}M_\odot$) within $0-0.5R_{500}$, as compared to a flat SMF for the low-mass clusters, suggesting the low-mass galaxies grow over time before reaching the cluster center. Additionally, we find a flatter SMF for the low redshift bin within $0.5R_{500}$ at stellar masses $M_*< 10^{10}M_\odot$. We also examined the effect of cluster ellipticity on the cluster galaxy SMF; however do not find statistically significant differences between the high and the low ellipticity clusters.

astro-ph.GA↗

arXiv2025

Towards Reasoning-Preserving Unlearning in Multimodal Large Language Models

Machine unlearning aims to erase requested data from trained models without full retraining. For Reasoning Multimodal Large Language Models (RMLLMs), this is uniquely challenging: intermediate chain-of-thought steps can still leak sensitive information even when final answers are forgotten, and overly aggressive interventions easily damage general reasoning ability. Yet no benchmark jointly evaluates how well unlearning methods suppress reasoning-level leakage while preserving reasoning competence. We address this gap with RMLLMU-Bench, the first benchmark for RMLLM unlearning that extends standard forgetting metrics with dedicated measures of reasoning leakage and reasoning retention. A systematic evaluation on RMLLMU-Bench reveals that existing unlearning methods for MLLMs and Large (Language) Reasoning Models (LRMs) either leave substantial leakage in the reasoning process or severely degrade reasoning performance. To address these gaps, we propose R-MUSE (Reasoning-preserving MLLM Unlearning via Subspace guidance and Adaptive Steering), a training-free and inference-time intervention framework that steers internal representations to forget both answers and reasoning traces while explicitly preserving general reasoning. Experiments on RMLLMU-Bench demonstrate that R-MUSE achieves a substantially better balance between effective forgetting and reasoning retention.

cs.CL↗

arXiv2025

Adaptive LLM Agents: Toward Personalized Empathetic Care

Current mental-health conversational systems are usually based on fixed, generic dialogue patterns. This paper proposes an adaptive framework based on large language models that aims to personalize therapeutic interaction according to a user's psychological state, quantified with the Acceptance of Illness Scale (AIS). The framework defines three specialized agents, L, M, and H, each linked to a different level of illness acceptance, and adjusts conversational behavior over time using continuous feedback signals. The AIS-stratified architecture is treated as a diegetic prototype placed in a plausible near-future setting and examined through the method of design fiction. By embedding the architecture in narrative scenarios, the study explores how such agents might influence access to care and therapeutic relationship. The goal is to show how clinically informed personalization, technical feasibility, and speculative scenario analysis can together inform the responsible design of LLM-based companions for mental-health support.

cs.HC↗

arXiv2025

PIXEL: Adaptive Steering Via Position-wise Injection with eXact Estimated Levels under Subspace Calibration

Reliable behavior control is central to deploying large language models (LLMs) on the web. Activation steering offers a tuning-free route to align attributes (e.g., truthfulness) that ensure trustworthy generation. Prevailing approaches rely on coarse heuristics and lack a principled account of where to steer and how strongly to intervene. To this end, we propose Position-wise Injection with eXact Estimated Levels (PIXEL), a position-wise activation steering framework that, in contrast to prior work, learns a property-aligned subspace from dual views (tail-averaged and end-token) and selects intervention strength via a constrained geometric objective with a closed-form solution, thereby adapting to token-level sensitivity without global hyperparameter tuning. PIXEL further performs sample-level orthogonal residual calibration to refine the global attribute direction and employs a lightweight position-scanning routine to identify receptive injection sites. We additionally provide representation-level guarantees for the minimal-intervention rule, supporting reliable alignment. Across diverse models and evaluation paradigms, PIXEL consistently improves attribute alignment while preserving model general capabilities, offering a practical and principled method for LLMs' controllable generation. Our code is available at https://github.com/V1centNevwake/PIXEL-Adaptive-Steering

cs.AI↗

arXiv2025

SGS: Segmentation-Guided Scoring for Global Scene Inconsistencies

We extend HAMMER, a state-of-the-art model for multimodal manipulation detection, to handle global scene inconsistencies such as foreground-background (FG-BG) mismatch. While HAMMER achieves strong performance on the DGM4 dataset, it consistently fails when the main subject is contextually misplaced into an implausible background. We diagnose this limitation as a combination of label-space bias, local attention focus, and spurious text-foreground alignment. To remedy this without retraining, we propose a lightweight segmentation-guided scoring (SGS) pipeline. SGS uses person/face segmentation masks to separate foreground and background regions, extracts embeddings with a joint vision-language model, and computes region-aware coherence scores. These scores are fused with HAMMER's original prediction to improve binary detection, grounding, and token-level explanations. SGS is inference-only, incurs negligible computational overhead, and significantly enhances robustness to global manipulations. This work demonstrates the importance of region-aware reasoning in multimodal disinformation detection. We release scripts for segmentation and scoring at https://github.com/Gaganx0/HAMMER-sgs

cs.CV↗

arXiv2025

DGM4+: Dataset Extension for Global Scene Inconsistency

The rapid advances in generative models have significantly lowered the barrier to producing convincing multimodal disinformation. Fabricated images and manipulated captions increasingly co-occur to create persuasive false narratives. While the Detecting and Grounding Multi-Modal Media Manipulation (DGM4) dataset established a foundation for research in this area, it is restricted to local manipulations such as face swaps, attribute edits, and caption changes. This leaves a critical gap: global inconsistencies, such as mismatched foregrounds and backgrounds, which are now prevalent in real-world forgeries. To address this, we extend DGM4 with 5,000 high-quality samples that introduce Foreground-Background (FG-BG) mismatches and their hybrids with text manipulations. Using OpenAI's gpt-image-1 and carefully designed prompts, we generate human-centric news-style images where authentic figures are placed into absurd or impossible backdrops (e.g., a teacher calmly addressing students on the surface of Mars). Captions are produced under three conditions: literal, text attribute, and text split, yielding three new manipulation categories: FG-BG, FG-BG+TA, and FG-BG+TS. Quality control pipelines enforce one-to-three visible faces, perceptual hash deduplication, OCR-based text scrubbing, and realistic headline length. By introducing global manipulations, our extension complements existing datasets, creating a benchmark DGM4+ that tests detectors on both local and global reasoning. This resource is intended to strengthen evaluation of multimodal models such as HAMMER, which currently struggle with FG-BG inconsistencies. We release our DGM4+ dataset and generation script at https://github.com/Gaganx0/DGM4plus

cs.CV↗

arXiv2025

Adaptive Token-Weighted Differential Privacy for LLMs: Not All Tokens Require Equal Protection

Large language models (LLMs) frequently memorize sensitive or personal information, raising significant privacy concerns. Existing variants of differential privacy stochastic gradient descent (DPSGD) inject uniform noise into every gradient step, significantly extending training time and reducing model accuracy. We propose that concentrating noise primarily on gradients associated with sensitive tokens can substantially decrease DP training time, strengthen the protection of sensitive information, and simultaneously preserve the model's performance on non-sensitive data. We operationalize this insight through Adaptive Token-Weighted Differential Privacy (ATDP), a modification of vanilla DP-SGD that adaptively assigns different gradient weights to sensitive and non-sensitive tokens. By employing a larger noise scale at the early stage of training, ATDP rapidly disrupts memorization of sensitive content. As a result, ATDP only requires a few additional epochs of lightweight post-processing following standard fine-tuning, injecting targeted noise primarily on parameters corresponding to sensitive tokens, thus minimally affecting the model's general capabilities. ATDP can be seamlessly integrated into any existing DP-based fine-tuning pipeline or directly applied to non-private models as a fast privacy-enhancing measure. Additionally, combined with an initial redacted fine-tuning phase, ATDP forms a streamlined DP pipeline that achieves comparable canary protection to state-of-the-art DP-SGD methods, significantly reduces the computational overhead of DP fine-tuning, shortening training time by approximately 90 percent, while achieving comparable or superior privacy protection and minimal accuracy degradation.

cs.LG↗

arXiv2025

From Prediction to Explanation: Multimodal, Explainable, and Interactive Deepfake Detection Framework for Non-Expert Users

The proliferation of deepfake technologies poses urgent challenges and serious risks to digital integrity, particularly within critical sectors such as forensics, journalism, and the legal system. While existing detection systems have made significant progress in classification accuracy, they typically function as black-box models, offering limited transparency and minimal support for human reasoning. This lack of interpretability hinders their usability in real-world decision-making contexts, especially for non-expert users. In this paper, we present DF-P2E (Deepfake: Prediction to Explanation), a novel multimodal framework that integrates visual, semantic, and narrative layers of explanation to make deepfake detection interpretable and accessible. The framework consists of three modular components: (1) a deepfake classifier with Grad-CAM-based saliency visualisation, (2) a visual captioning module that generates natural language summaries of manipulated regions, and (3) a narrative refinement module that uses a fine-tuned Large Language Model (LLM) to produce context-aware, user-sensitive explanations. We instantiate and evaluate the framework on the DF40 benchmark, the most diverse deepfake dataset to date. Experiments demonstrate that our system achieves competitive detection performance while providing high-quality explanations aligned with Grad-CAM activations. By unifying prediction and explanation in a coherent, human-aligned pipeline, this work offers a scalable approach to interpretable deepfake detection, advancing the broader vision of trustworthy and transparent AI systems in adversarial media environments.

cs.CV↗

arXiv2025

eX-NIDS: A Framework for Explainable Network Intrusion Detection Leveraging Large Language Models

This paper introduces eX-NIDS, a framework designed to enhance interpretability in flow-based Network Intrusion Detection Systems (NIDS) by leveraging Large Language Models (LLMs). In our proposed framework, flows labelled as malicious by NIDS are initially processed through a module called the Prompt Augmenter. This module extracts contextual information and Cyber Threat Intelligence (CTI)-related knowledge from these flows. This enriched, context-specific data is then integrated with an input prompt for an LLM, enabling it to generate detailed explanations and interpretations of why the flow was identified as malicious by NIDS. We compare the generated interpretations against a Basic-Prompt Explainer baseline, which does not incorporate any contextual information into the LLM's input prompt. Our framework is quantitatively evaluated using the Llama 3 and GPT-4 models, employing a novel evaluation method tailored for natural language explanations, focusing on their correctness and consistency. The results demonstrate that augmented LLMs can produce accurate and consistent explanations, serving as valuable complementary tools in NIDS to explain the classification of malicious flows. The use of augmented prompts enhances performance by over 20% compared to the Basic-Prompt Explainer.

cs.CR↗

arXiv2025

Representation Learning on Out of Distribution in Tabular Data

The open-world assumption in model development suggests that a model might lack sufficient information to adequately handle data that is entirely distinct or out of distribution (OOD). While deep learning methods have shown promising results in handling OOD data through generalization techniques, they often require specialized hardware that may not be accessible to all users. We present TCL, a lightweight yet effective solution that operates efficiently on standard CPU hardware. Our approach adapts contrastive learning principles specifically for tabular data structures, incorporating full matrix augmentation and simplified loss calculation. Through comprehensive experiments across 10 diverse datasets, we demonstrate that TCL outperforms existing models, including FT-Transformer and ResNet, particularly in classification tasks, while maintaining competitive performance in regression problems. TCL achieves these results with significantly reduced computational requirements, making it accessible to users with limited hardware capabilities. This study also provides practical guidance for detecting and evaluating OOD data through straightforward experiments and visualizations. Our findings show that TCL offers a promising balance between performance and efficiency in handling OOD prediction tasks, which is particularly beneficial for general machine learning practitioners working with computational constraints.

cs.LG↗