arXiv subjects

Prabhjot Singh

Publications and source records attributed to Prabhjot Singh.

At least 19 recordsLinked to original sources

arXiv2026

Not Truly Multilingual: Script Consistency as a Missing Dimension in VLM Evaluation

Current multilingual evaluations for Vision-Language Models (VLMs) assume a one-to-one mapping between language and orthography, overlooking billions of users of multi-script languages. We introduce PuMVR (Punjabi Multimodal Visual Reasoning), a benchmark of 1,000 strictly parallel image-text instances across Punjabi's three active scripts: Gurmukhi, Shahmukhi, and Roman. Evaluating 10 state-of-the-art VLMs, we expose a substantial and systematic Script Gap. Models frequently solve visual tasks in one script while failing identical tasks in another, with accuracy deltas reaching 16%. Crucially, visual input boosts absolute performance uniformly yet does not close the orthographic gap. Furthermore, cross-script in-context transfer is highly brittle, exposing script-locked knowledge representation. Supported by McNemar tests across all script pairs, our findings demonstrate that current "multilingual" VLMs are not truly multi-script. We propose the Script Consistency Rate (SCR), which falls as low as 24.8% on our benchmark, as a mandatory metric for script-agnostic evaluation to ensure equitable AI access. Data and code are available at: https://github.com/prabhjotschugh/Not-Truly-Multilingual-PuMVR.

cs.CV↗

arXiv2026

When Retrieval Helps and Distracts: Evaluating Evidence-Generating LLMs for Biomedical Claim Verification

Biomedical fact-checking systems must do more than predict whether a claim is supported, contradicted, or unaddressed: they should also produce evidence that is faithful, complete, and useful for verification. We study this evidence-generation setting on CARE-XAI, a unified benchmark spanning five biomedical and health fact-checking sources. We compare base instruction LLMs, PubMed retrieval-augmented LLMs, fine-tuned LLMs, label-only LLMs, and biomedical encoder classifiers under a shared evaluation protocol. Biomedical classifiers remain strongest for verdict-only prediction, while fine-tuned LLMs are the strongest evidence-generating systems. PubMed retrieval is mixed: it helps PubMed-aligned sources such as PubMedQA and SciFact, but can distract models on broader public-health claims. We introduce Bio-GRACE, a gold-reference-normalized diagnostic for measuring whether retrieved evidence recovers the decision benefit of reference evidence. Bio-GRACE shows that retrieval utility is source-dependent, motivates selective retrieval, and exposes why retrieval recall and lexical evidence overlap are insufficient for biomedical fact-checking.

cs.CL↗

arXiv2026

AFDBench: A Reasoning-First AI Scientist for NationalWeather Service Forecast Discussions

Large language models (LLMs) hallucinate numerical values when generating high-stakes meteorological text, posing risks for weather communication. We present AFDBench, an AI meteorologist that generates professional Area Forecast Discussions (AFDs) by reasoning through structured AI weather forecast data from Google's WeatherNext 2. We introduce AFDBench, the first benchmark for evaluating generative meteorological reasoning, comprising 7,732 expert written discussions from 13 National Weather Service (NWS) offices paired with real AI weather forecast inputs, and three complementary metrics: Met-Align (numerical accuracy), Style-Align (professional dialect adherence), and Input-Grounding (fidelity to source weather data). Zero-shot evaluations reveal that open-source LLMs achieve low Style-Align (~0.33) and moderate Input-Grounding (~0.88), failing to write in the professional NWS register or faithfully use their input data. We apply Group Relative Policy Optimization (GRPO) with domain-specific rewards targeting temperature accuracy, synoptic correctness, and format compliance. On 1,033 held-out samples from two unseen NWS offices, GRPO nearly doubles Style-Align from 0.318 to 0.619 and improves Input-Grounding from 0.881 to 0.940, demonstrating that reinforcement learning teaches a 7B-parameter model to write like a professional meteorologist and faithfully interpret AI weather data.

cs.LG↗

arXiv2026

How Much Reconstruction Does Quantum Machine Learning Need? Late Fusion of Independently Trained Quantum Subcircuits

Circuit cutting lets a large quantum neural network (QNN) run as independent subcircuits on small devices, but rebuilding its outputs by reconstruction carries a classical sampling overhead exponential in the number of cuts - the dominant runtime cost in prior work. We ask whether, for machine-learning tasks, this step is necessary, and replace it with late fusion: each subcircuit is trained and measured independently, and a small classical head combines their outputs - a linear-cost, decision-level combination borrowed from multimodal learning. To characterize the trade-off we introduce a quantumness dial $Q$, a tunable reconstruction budget interpolating from pure fusion to full reconstruction, and a cut-entanglement diagnostic that indicates how much reconstruction a task needs (Spearman $ρ=0.59$ over $104$ runs). Across synthetic and standard datasets, independently trained late fusion matches full reconstruction accuracy within $0.04$ at every point of the controlled sweep and on every classical benchmark, at exponentially lower cost; it is also markedly more robust to shot and device noise. Controlled entangled-data experiments locate the boundary where fusion must fail. We do not claim advantage over classical machine learning - consistent with recent benchmarking, quantum offers no accuracy edge on these datasets. Late fusion is thus an efficient, noise-robust, self-characterizing alternative to reconstruction for circuit-cutting QML.

quant-ph↗

arXiv2026

Same Facts, Different Diagnosis: Measuring and Mitigating Narrative Anchoring in Clinical Language Models

Large language models used for clinical diagnostic reasoning are sensitive to sociolinguistic register, not just clinical content. We term this failure mode Narrative Anchoring: identical clinical facts expressed in different registers cause diagnostic outputs to diverge. Unlike prior demographic-bias work, which manipulates explicit identity tokens such as race or income, our benchmark isolates register as the sole channel of variation, with no demographic marker present in any form. We construct a dataset of 1,000 USMLE clinical vignettes, each rewritten into three sociolinguistically distinct personas under an independently audited fact-preservation guarantee, verified by a separate model that never sees the generation prompt. Across seven language models spanning three architecture families and scales, Narrative Anchoring is statistically significant under direct prompting in every model tested, with a Narrative Anchoring Gap of 0.064 to 0.151. Chain-of-thought reasoning and explicit debiasing instructions reduce the bias only partially, and their apparent gains are frequently confounded by accuracy collapse. We introduce NarrativeShield, a three-agent pipeline that structurally extracts and verifies clinical facts before diagnostic reasoning begins, reducing the Narrative Anchoring Gap to near-zero ($-0.004$ to $0.037$) and achieving the lowest rate of severely unstable decisions (DSS $<$ 0.8) of any method across all models, at a modest and mechanistically expected accuracy cost for most models. A stress test using a non-instruction-tuned base model shows that executing a debiasing intervention at all is gated by zero-shot instruction-following ability, not prompt content alone. We release our dataset, human-validated for fact preservation, as a standalone resource for studying register-based clinical bias.

cs.CL↗

arXiv2026

ReVision : A Post-Hoc, Vision-Based Technique for Replacing Unacceptable Concepts in Image Generation Pipeline

Image-generative models are widely deployed across industries. Recent studies show that they can be exploited to produce unacceptable content. Existing mitigation strategies rely on prompt filtering and safety-aware training, both of which can be bypassed and often degrade generative quality. In this work, we propose ReVision, a training-free, prompt-based, post-hoc safety framework for image-generation pipeline. ReVision acts as a post-generation safeguard by analyzing generated images and selectively editing unsafe concepts without altering the underlying generator. Prior post-hoc editing methods often rely on imprecise spatial localization, limiting deployability, in multi-concept scenes. To address this limitation, ReVision introduces a VLM-assisted spatial gating mechanism for instance-consistent localization, enabling integrity-preserving edits. We introduce an 800-image benchmark spanning single- and multi-unsafe-concept images, each composed alongside benign concepts in shared scenes. On this benchmark, ReVision improves CLIP alignment toward safe prompts by +0.121, reduces multi-concept background LPIPS from 0.166 to 0.058, and eliminates NudeNet detections (70.51 -> 0). Across external benchmarks, ReVision outperforms prior methods, and a human study shows it reduces recognizability of unacceptable content from 96% to 10%.

cs.CR↗

arXiv2026

DocHRL: A Hierarchical Reinforcement Learning Framework for Cost-Optimised Document Classification

Real-world document classification pipelines typically apply the same sequence of models to every incoming document, regardless of its complexity or type. This leads to inefficient use of compute and human resources: simple documents are over-processed while difficult ones may not receive enough scrutiny. We introduce DocHRL, a hierarchical reinforcement learning framework that learns to adaptively and dynamically select the most cost-effective classification policy on a per-document basis. DocHRL formulates document classification as a sequential decision problem with a two-level policy hierarchy: a top-level policy selects among broad options (vision classifiers, LLMs, OCR, and human-in-the-loop review), while option-specific sub-policies choose the concrete model or tool to invoke. The reward signal is the negative total expected cost, which captures inference cost, cost of misclassification, and cost of human labelling. Trained with Proximal Policy Optimisation on the RVL-CDIP benchmark, DocHRL achieves a macro F1 of 0.973 across 16 document classes while reducing average per-document cost to 2.74 normalised units compared to substantially higher costs incurred by fixed standalone classifiers. Our results demonstrate that cost-aware reinforcement learning can simultaneously improve classification performance and operational efficiency in document understanding systems.

cs.AI↗

arXiv2026

FIRSTPASS: A Multi-Domain, Multi-Round Peer Review Dataset Grounded in Real Editorial Outcomes

Scientific peer review datasets have trained AI systems exclusively on Computer Science and Machine Learning venues, producing models that critique ablation studies yet have never seen a biology reviewer demand contamination controls or a chemist question Nuclear Magnetic Resonance (NMR) spectral assignments. We introduce FIRSTPASS, the first large-scale peer review dataset built on complete multi-round editorial dialogues from a multidisciplinary high-impact journal. Curated from Nature Communications mandatory transparent peer review (instituted November 2022), FIRSTPASS comprises 3,668 records spanning five scientific domains (biology, chemistry, neuroscience, physics, and earth science), capturing the full iterative structure of scientific validation: initial referee reports, author point-by-point responses, and updated reviewer assessments. Each record carries an outcome label derived directly from editorial decisions (STANDARD for two-round review; EXTENDED for three or more rounds), providing ground truth absent in all prior corpora. An automated audit confirms 100% content integrity. Expert reviews average 2,155 words, substantially denser than conference venue reviews. All data, parsing pipelines, and evaluation scripts are released to enable reproducible benchmarking of AI scientific judgment across disciplines.

cs.CL↗

arXiv2026

The Hallucination Snowball: Modeling Error Propagation as State Transitions in Multi-Agent LLM Pipelines

Sequential multi-agent LLM pipelines chain specialized agents without verification at handoffs, creating a structural flaw with measurable and severe consequences. We show that hallucinations injected at Stage 1 do not merely persist; they transform: raw numerical facts become derived computations, then narrative prose, then editorially approved conclusions. At each transformation, detectability degrades near-irreversibly. We formalize this as the hallucination snowball effect, a first-order Markov process over four states (Raw Fact $\to$ Derived $\to$ Narrative $\to$ Invisible) with empirically measured per-boundary escape probabilities of 24.6%, 48.3%, and 89.3%. Across 346 automatically injected hallucinations in a 4-agent financial analysis pipeline on FinanceBench, gpt-4o detection drops from 72.0% at Stage 1 to 50.9% at Stage 4, and 23.7% of hallucinations survive completely undetected in the final output. Even the strongest model tested (Qwen3.5-397B-A17B, 87.0% at Stage 1) faces a structural ceiling; projected Stage 4 detection is only ${\sim}$60--65%. Critically, boundary gates using identical RAG verification tools reduce hallucination survival from 58.4% to 16.2% versus end-of-pipeline checking (Cohen's $h = -0.911$, $p < 0.000001$), while end-checking alone achieves merely 2.3 pp improvement over no verification. When you verify matters more than whether you verify. Our model predicts survival for $n$-agent linear pipelines and prescribes optimal verification resource allocation: invest at $S_1{\to}S_2$ first, where 75.4% of hallucinations are still catchable, not at $S_3{\to}S_4$ where 89.3% have already escaped.

cs.AI↗

arXiv2026

DistributedEstimator: Distributed Training of Quantum Neural Networks via Circuit Cutting

Circuit cutting decomposes a large quantum circuit into smaller subcircuits executed independently; expectation values are recovered by classically combining subcircuit outcomes. Prior work characterises cutting overhead via subcircuit counts and sampling complexity, but its end-to-end impact on iterative, estimator-driven training pipelines remains under-measured from a systems perspective. We propose DistributedEstimator, a cut-aware estimator execution pipeline that treats circuit cutting as a staged distributed workload, instrumenting each query across four phases: partitioning, subexperiment generation, parallel execution, and classical reconstruction. Using logged runtime traces and learning outcomes on two binary classification workloads (Iris and MNIST), we quantify cutting overheads, scaling limits, and sensitivity to injected stragglers, and assess whether accuracy and robustness are preserved under matched training budgets. Reconstruction dominates per-query time -- a median of 53% and 95th percentile of 58% at three cuts -- bounding achievable speed-up under parallelism. Despite this, test accuracy is fully preserved on Iris and maintained without systematic degradation on MNIST across all cut configurations. Robustness under Gaussian noise and FGSM perturbations is similarly preserved, with several configurations matching or improving on the uncut baseline. Exponential growth of subexperiment counts (${O}(9^c)$ for CNOT-based decomposition) is a fundamental barrier limiting practical experimentation to small qubit counts. These results establish that practical scaling for learning workloads requires reducing and overlapping reconstruction, scheduling policies for barrier-dominated critical paths, and computationally efficient reconstruction strategies for larger qubit counts.

cs.DC↗

arXiv2026

FirstPass: Grounding AI Scientific Judgment in Multi-Round Editorial Outcomes

AI systems for peer review fail on three fronts: they train on Computer Science and Machine Learning venues alone, ignore the iterative dialogue that validates science, and evaluate on stylistic mimicry rather than real editorial judgment. We introduce FirstPass, a dataset and fine-tuned model that addresses all three. Curating 3,668 complete multi-round peer-review dialogues from Nature Communications across five scientific domains (biology, chemistry, neuroscience, physics, and earth science), we exploit mandatory transparent peer review (instituted November 2022) and verify 100% content integrity by automated audit. We fine-tune Qwen2.5-7B-Instruct via Low-Rank Adaptation (LoRA) on three tasks: review generation, reviewer updating, and revision-cycle prediction. Our key finding is that response-only loss masking is a prerequisite, not an optimization: without it, accuracy is 62.0%, below the majority baseline; with it, FirstPass achieves 80.5% accuracy and F1-macro 78.2% on predicting editorial outcomes (Standard vs. Extended revision cycles), outperforming Gemini-3.1-flash-lite-preview zero-shot by 10.4 percentage points and all baselines with statistical significance (McNemar p < 0.001). On generation, FirstPass produces reviews averaging 1,187 words, substantially closer to human references (2,155 words) than any baseline, achieving ROUGE-L 0.154 with significant gains over Qwen and DeepSeek zero-shot (p < 0.001). Deployed in the pre-submission loop as an anticipatory scientific co-author, FirstPass simulates expert critique and predicts revision cycle outcomes before submission, giving authors the judgment a trusted colleague would provide, with consistent cross-domain performance across five disciplines.

cs.CL↗

arXiv2026

Beyond 'One Language, One Script': Quantifying Orthographic Bias in Multilingual VLMs with PuMVR

Current Vision-Language Models (VLMs) are celebrated for their multilingual capabilities, yet they operate under a flawed assumption: that one language corresponds to a single writing system. This overlooks billions of users of multi-script languages like Punjabi, Serbian, Hindi-Urdu, Kurdish, among many others, for whom a model's capability may be fractured by orthographic bias. We introduce PuMVR (Punjabi Multimodal Visual Reasoning), the first benchmark designed to quantify script-dependent bias through 375 culturally grounded image-reasoning tasks across Punjabi's three active scripts (Gurmukhi, Shahmukhi, Roman). Evaluating 10 state-of-the-art VLMs, we expose a substantial Script Gap: models frequently solve visual puzzles in one script while failing identical tasks in another, with accuracy deltas reaching 16% and Script Consistency Rates (SCR) as low as 24.8%. Crucially, visual input boosts absolute performance but does not close this gap, the relative bias persists. Our analysis suggests reasoning patterns show limited cross-script transferability, and Chain-of-Thought pathways diverge based on script alone. We propose SCR as a core metric for script-agnostic evaluation, challenging current multilingual assessment paradigms and providing a framework for equitable AI.

cs.CL↗

arXiv2026

AirCast-SR: A Foundation Model for Kilometer-Scale Atmospheric Super-Resolution via Latent Consistency Diffusion

Operational weather prediction at kilometer scales remains computationally prohibitive for traditional numerical weather prediction (NWP) models, limiting forecast access for applications in energy, agriculture, and disaster management that require fine-grained spatiotemporal detail. Here we introduce AirCast-SR, a foundation model for atmospheric super-resolution that downscales global AI weather forecasts from 0.25 degree (~28 km) to 1 km horizontal resolution at hourly temporal resolution, producing 67-hour forecasts of eight coupled surface variables simultaneously. EarthMind-SR employs a three-dimensional U-Net conditioned within a Latent Consistency Model (LCM) diffusion framework, trained on patch-based samples over the contiguous United States (CONUS) using GraphCast forecasts as input and NOAA's Analysis of Record for Calibration (AORC) as the target. The model achieves near-zero bias across all variables and lead times, and its radial power spectral density analysis demonstrates preservation of fine-scale atmospheric structure at wavelengths of 10 km to 100 km where coarser models lose spectral power. We validate EarthMind-SR across three CONUS case studies spanning winter, summer, and spring seasons, and demonstrate zero-shot global transferability over India and Germany using independent surface station observations without any retraining or fine-tuning. As an open-weights foundation model, EarthMind-SR establishes a new paradigm for kilometer-scale AI weather prediction and provides a platform for regional fine-tuning, distillation, and downstream applications in climate services and hazard forecasting.

cs.LG↗

arXiv2026

Learning from Disagreement: Clinician Overrides as Implicit Preference Signals for Clinical AI in Value-Based Care

We reframe clinician overrides of clinical AI recommendations as implicit preference data - the same signal structure exploited by reinforcement learning from human feedback (RLHF), but richer: the annotator is a domain expert, the alternatives carry real consequences, and downstream outcomes are observable. We present a formal framework extending standard preference learning with three contributions: a five-category override taxonomy mapping override types to distinct model update targets; a preference formulation conditioned on patient state s, organizational context c, and clinician capability kappa, where kappa decomposes into execution capability kappa-exec and alignment capability kappa-align; and a dual learning architecture that jointly trains a reward model and a capability model via alternating optimization, preventing a failure mode we term suppression bias-the systematic suppression of correct-but-difficult recommendations when clinician capability falls below the execution threshold. We argue that chronic disease management under outcome-based payment contracts produces override data with uniquely favorable properties-longitudinal density, concentrated decision space, outcome labels, and natural capability variation-and that training environments combining longitudinal outcome measurement with aligned financial incentives are a necessary condition for learning a reward model aligned with patient trajectory rather than with encounter economics. This framework emerged from operational work to improve clinician capability in a live value-based care deployment.

cs.LG↗

arXiv2026

Learning to Compress Time-to-Control: A Reinforcement Learning Framework for Chronic Disease Management

Reinforcement learning (RL) in healthcare has had mixed results, with reward sparsity, unreliable off-policy evaluation, and deployment-simulation gap as recurring failure modes. We argue that chronic disease management is structurally a more tractable RL setting than the acute-care problems the field has primarily studied, but only if the problem is formalized to exploit chronic care's properties. We propose such a formalization. The agent's objective is to compress time-to-control (TTC) under a tiered reward calibrated to the CMS ACCESS Model. Two quantities from our companion preference-learning paper [Singh et al. 2026] enter as load-bearing structural elements: the execution intensity εbounds action availability under a constrained Markov Decision Process, and the clinician capability κweights offline-data transitions during RL training. Together they couple preference learning and RL into a two-loop architecture. We present simulation results on synthetic state machines for hypertension and type 2 diabetes. Capability-weighted offline RL outperforms uniform-weighted offline RL and the behavior policy by 15 percentage points on T2D TTC; the uniform-weighted formulation (the standard in existing healthcare RL) underperforms even the heterogeneous behavior policy. \Epsilon-aware policies generalize across deployment regimes while ε-naive policies do not.

cs.LG↗

arXiv2026

When Less Is More: Simplicity Beats Complexity for Physics-Constrained InSAR Phase Unwrapping

Operational phase unwrapping is the primary computational bottleneck in InSAR-based volcanic and seismic monitoring. We challenge the industry trend of adopting high-complexity computer vision architectures, such as attention mechanisms, without validating their suitability for physics-constrained geophysical regression. We present the first large-scale architectural ablation study on a global LiCSAR benchmark (20 frames, 39,724 patches, 651M pixels). Our results reveal a significant "complexity penalty": a vanilla U-Net (7.76M parameters) achieves $R^2=0.834$ and RMSE $= 1.01$ cm, outperforming 11.37M-parameter attention-based models by 34% in $R^2$ and 51% in RMSE. Power Spectral Density (PSD) analysis provides the physical justification: while attention excels at capturing sharp semantic edges in natural images, it injects unphysical high-frequency artifacts ($>0.3$ cycles/pixel) into geophysical fields, violating the fundamental smoothness constraints of elastic surface deformation. With a 2.92ms inference latency (a $2.5\times$ speedup), the vanilla U-Net is the only candidate to comfortably meet the sub-100ms requirement for operational early-warning systems. This work bridges the "publication-to-practice" gap by proving that convolutional locality outperforms modern complexity for smooth-field regression, advocating for physics-informed simplicity in ML4RS. Code available at https://github.com/prabhjotschugh/When-Less-is-More-InSAR-Phase-Unwrapping

cs.CV↗

arXiv2025

Quantum Federated Learning: Architectural Elements and Future Directions

Federated learning (FL) focuses on collaborative model training without the need to move the private data silos to a central server. Despite its several benefits, the classical FL is plagued with several limitations, such as high computational power required for model training(which is critical for low-resource clients), privacy risks, large update traffic, and non-IID heterogeneity. This chapter surveys a hybrid paradigm - Quantum Federated Learning (QFL), which introduces quantum computation, that addresses multiple challenges of classical FL and offers rapid computing capability while keeping the classical orchestration intact. Firstly, we motivate QFL with a concrete presentation on pain points of classical FL, followed by a discussion on a general architecture of QFL frameworks specifying the roles of client and server, communication primitives and the quantum model placement. We classify the existing QFL systems based on four criteria - quantum architecture (pure QFL, hybrid QFL), data processing method (quantum data encoding, quantum feature mapping, and quantum feature selection & dimensionality reduction), network topology (centralized, hierarchial, decentralized), and quantum security mechanisms (quantum key distribution, quantum homomorphic encryption, quantum differential privacy, blind quantum computing). We then describe applications of QFL in healthcare, vehicular networks, wireless networks, and network security, clearly highlighting where QFL improves communication efficiency, security, and performance compared to classical FL. We close with multiple challenges and future works in QFL, including extension of QFL beyond classification tasks, adversarial attacks, realistic hardware deployment, quantum communication protocols deployment, aggregation of different quantum models, and quantum split learning as an alternative to QFL.

quant-ph↗

arXiv2025

Advanced Real-Time Fraud Detection Using RAG-Based LLMs

Artificial Intelligence has become a double edged sword in modern society being both a boon and a bane. While it empowers individuals it also enables malicious actors to perpetrate scams such as fraudulent phone calls and user impersonations. This growing threat necessitates a robust system to protect individuals In this paper we introduce a novel real time fraud detection mechanism using Retrieval Augmented Generation technology to address this challenge on two fronts. First our system incorporates a continuously updating policy checking feature that transcribes phone calls in real time and uses RAG based models to verify that the caller is not soliciting private information thus ensuring transparency and the authenticity of the conversation. Second we implement a real time user impersonation check with a two step verification process to confirm the callers identity ensuring accountability. A key innovation of our system is the ability to update policies without retraining the entire model enhancing its adaptability. We validated our RAG based approach using synthetic call recordings achieving an accuracy of 97.98 percent and an F1score of 97.44 percent with 100 calls outperforming state of the art methods. This robust and flexible fraud detection system is well suited for real world deployment.

cs.CR↗