arXiv subjects

Fahim Ahmed

Publications and source records attributed to Fahim Ahmed.

10 recordsLinked to original sources

arXiv2026

Long-range scalar interactions in the effective field theory approach to $0νββ$

As a lepton-number-violating process whose observation would establish the Majorana nature of neutrinos, neutrinoless double-beta decay ($0νββ$) is an important low-energy nuclear probe of beyond-the-Standard-Model physics. Effective field theory (EFT) provides a general, model-independent description, and interference among effective mechanisms is essential for disentangling their contributions to the total decay rate. We focus on scalar-type terms of the long-range Lagrangian in the EFT approach to $0νββ$. We simultaneously retain the two scalar interactions involving a common $(S+P)$ leptonic current coupled either to an $(S-P)$ or an $(S+P)$ hadronic current. Their coherent contribution produces mutual interference between the two scalar mechanisms, controlled by the relative complex phase of their couplings, in addition to their separate interference with the standard mass mechanism. We give a compact half-life expression, the explicit scalar--scalar interference coefficient in terms of scalar nuclear-matrix-element ratios and integrated phase-space factors, and a positivity bound on that coefficient. The calculation uses the closure and single-nucleon impulse approximations, retains the dominant scalar recoil subset, and includes the $s_{1/2}$ and $p_{1/2}$ Coulomb waves of the outgoing electrons. The resulting dimensionally explicit formulation provides a basis for correlated analyses of the two scalar coefficients.

hep-ph↗

arXiv2026

Khondo: A Multimodal Benchmark for Document Packet Splitting of Bangla Forms

Document packets, multiple documents concatenated into a single file, are common in government and administrative workflows, yet splitting them into their constituent documents is difficult, especially for low-resource languages. We introduce Khondo (Bangla for split/segment), the first benchmark for document packet splitting on Bangladeshi government forms. Unlike prior English and OCR-text-based datasets, Khondo is bilingual (Bangla--English) and vision-native; where models operate directly on page images. It spans five concatenation schemes, from sequential to fully shuffled, across 14 administrative domains, with ground-truth boundaries, domain types, and page order. Zero-shot evaluation of MLLMs shows they cluster pages into their source documents fairly well but struggle in restoring the original page order once shuffled. To isolate what drives this difficulty, we run two controlled analyses, varying the prompt instruction and then the packet language. Both primarily affect ordering rather than clustering: (a) explicit page-order instructions are necessary but insufficient, and (b) English packets are ordered more reliably than Bangla, making page arrangement the dominant challenge and language a secondary but consistent factor. Khondo establishes page-order reconstruction as a key open problem in vision-based, low-resource document understanding, and provides a controlled benchmark for measuring progress toward solving it. Our dataset and code is available at https://huggingface.co/datasets/Mausul/khondo

cs.CL↗

arXiv2026

How Analysts Use AI in High-Stakes Crime Linkage: An Industrial Study

Crime linkage analysis is used in many countries to identify series of offences that may have been committed by the same individual. In practice, specialist analysts manually search for behavioural and situational connections across large crime databases, an effort that is time-consuming, cognitively demanding, and can involve repeated exposure to disturbing material. To support this work, an Artificial Intelligence (AI)-enabled decision-support tool was co-developed with a UK law enforcement agency to assist analysts in identifying likely crime linkages. This paper reports an industrial evaluation of the crime-linkage tool. We conducted a mixed-methods usability study combining direct observation, eye-tracking, mouse-tracking, and surveys to examine how analysts engage with AI predictions and with the model features presented as explanations. Our findings show that analysts used the AI predictions selectively and frequently validated them against behavioural (non-AI) evidence, reflecting partial trust and an ongoing reliance on established analytical practices. We also found that analysts attended to the presented model features and valued their availability, while identifying opportunities to improve how explanations are presented and integrated into the workflow. Overall, our results highlight the need for AI-enabled decision-support tools to better integrate explanations and traditional analytical methods, and demonstrate the importance of in-situ evaluation for engineering usable and trustworthy AI in high-stakes settings.

cs.HC↗

arXiv2026

BaFCo: A Document Understanding Benchmark for Complex Bangla Form Comprehension

Document comprehension is a challenging yet impactful task for Multimodal Large Language Models, especially as these systems see growing adoption in real-world, human-centric applications. However, this adoption is limited for low-resource languages such as Bangla due to the scarcity of high-quality annotated data. To address this gap, we introduce BaFCo, a benchmark dataset for Bangla form comprehension with a focus on Document Layout Analysis (DLA) and Key Information Extraction (KIE). BaFCo curates 200 multi-page complex Bangladeshi government forms, sourced from across diverse sectors including agriculture, education, banking, and land management. To accurately capture the structural and contextual complexity of these forms, we define a fine-grained annotation schema comprising 26 types of form entities, along with a separate coarse form entity set consisting of 5 types. We evaluate the latest MLLMs from the ChatGPT, Gemini, Claude, Qwen, and Kimi series using zero-shot and chain-of-thought prompts under both low and high reasoning setups. Our results reveal limitations in current MLLMs' ability in comprehending Bangla forms, particularly in accurately localizing highly granular form entities. Our dataset and code is available at: https://huggingface.co/datasets/Mausul/bafco

cs.CL↗

arXiv2026

Enhancing Binary Encoded Crime Linkage Analysis Using Siamese Network

Effective crime linkage analysis is crucial for identifying serial offenders and enhancing public safety. To address limitations of traditional crime linkage methods in handling high-dimensional, sparse, and heterogeneous data, we propose a Siamese Autoencoder framework that learns meaningful latent representations and uncovers correlations in complex crime data. Using data from the Violent Crime Linkage Analysis System (ViCLAS), maintained by the Serious Crime Analysis Section of the UK's National Crime Agency, our approach mitigates signal dilution in sparse feature spaces by integrating geographic-temporal features at the decoder stage. This design amplifies behavioral representations rather than allowing them to be overshadowed at the input level, yielding consistent improvements across multiple evaluation metrics. We further analyze how different domain-informed data reduction strategies influence model performance, providing practical guidance for preprocessing in crime linkage contexts. Our results show that advanced machine learning approaches can substantially enhance linkage accuracy, improving AUC by up to 9% over traditional methods while offering interpretable insights to support investigative decision-making.

cs.LG↗

arXiv2025

BAPPA: Benchmarking Agents, Plans, and Pipelines for Automated Text-to-SQL Generation

Text-to-SQL systems provide a natural language interface that can enable even laymen to access information stored in databases. However, existing Large Language Models (LLM) struggle with SQL generation from natural instructions due to large schema sizes and complex reasoning. Prior work often focuses on complex, somewhat impractical pipelines using flagship models, while smaller, efficient models remain overlooked. In this work, we explore three multi-agent LLM pipelines, with systematic performance benchmarking across a range of small to large open-source models: (1) Multi-agent discussion pipeline, where agents iteratively critique and refine SQL queries, and a judge synthesizes the final answer; (2) Planner-Coder pipeline, where a thinking model planner generates stepwise SQL generation plans and a coder synthesizes queries; and (3) Coder-Aggregator pipeline, where multiple coders independently generate SQL queries, and a reasoning agent selects the best query. Experiments on the Bird-Bench Mini-Dev set reveal that Multi-Agent discussion can improve small model performance, with up to 10.6% increase in Execution Accuracy for Qwen2.5-7b-Instruct seen after three rounds of discussion. Among the pipelines, the LLM Reasoner-Coder pipeline yields the best results, with DeepSeek-R1-32B and QwQ-32B planners boosting Gemma 3 27B IT accuracy from 52.4% to the highest score of 56.4%. Codes are available at https://github.com/treeDweller98/bappa-sql.

cs.CL↗

arXiv2024

Assignment of Freight Traffic in a Large-Scale Intermodal Network under Uncertainty

This paper presents a methodology for freight traffic assignment in a large-scale road-rail intermodal network under uncertainty. Network uncertainties caused by natural disasters have dramatically increased in recent years. Several of these disasters (e.g., Hurricane Sandy, Mississippi River Flooding, and Hurricane Harvey) severely disrupted the U.S. freight transportation network, and consequently, the supply chain. To account for these network uncertainties, a stochastic freight traffic assignment model is formulated. An algorithmic framework, involving the sample average approximation and gradient projection algorithm, is proposed to solve this challenging problem. The developed methodology is tested on the U.S. intermodal network with freight flow data from the Freight Analysis Framework. The experiments consider three types of natural disasters that have different risks and impacts on transportation networks: earthquakes, hurricanes, and floods. It is found that for all disaster scenarios, freight ton-miles are higher compared to the base case without uncertainty. The increase in freight ton-miles is the highest under the flooding scenario; this is because there are more states in the flood-risk areas, and they are scattered throughout the U.S.

math.OC↗

arXiv2021

Hybrid Classical-Quantum Deep Learning Models for Autonomous Vehicle Traffic Image Classification Under Adversarial Attack

Image classification must work for autonomous vehicles (AV) operating on public roads, and actions performed based on image misclassification can have serious consequences. Traffic sign images can be misclassified by an adversarial attack on machine learning models used by AVs for traffic sign recognition. To make classification models resilient against adversarial attacks, we used a hybrid deep-learning model with both the quantum and classical layers. Our goal is to study the hybrid deep-learning architecture for classical-quantum transfer learning models to support the current era of intermediate-scale quantum technology. We have evaluated the impacts of various white box adversarial attacks on these hybrid models. The classical part of hybrid models includes a convolution network from the pre-trained Resnet18 model, which extracts informative features from a high dimensional LISA traffic sign image dataset. The output from the classical processor is processed further through the quantum layer, which is composed of various quantum gates and provides support to various quantum mechanical features like entanglement and superposition. We have tested multiple combinations of quantum circuits to provide better classification accuracy with decreasing training data and found better resiliency for our hybrid classical-quantum deep learning model during attacks compared to the classical-only machine learning models.

quant-ph↗

arXiv2020

Interference Effects for $0νββ$ Decay in the Left-Right Symmetric Model

Various mechanisms may contribute to neutrinoless double beta decay in the left-right symmetric model. The interference between these mechanisms also contribute to the overall decay rate. The analysis of the contributions of these interference terms is important for disentangling different mechanisms. In the present paper we study interference effects contributing to the decay rate for neutrinoless double-$β$ decay in the left-right symmetric model. The numerical values for maximum interference for several nuclides are calculated. It is observed that, for most of the interference terms, the contribution is smaller than 20$\%$ for all the nuclei considered in the study. However, the interference between the mass mechanisms (light and heavy) and $η$ mechanism is observed to be in the range 30$\%$-50$\%$. The variation of the interference effect with the $Q$ values is also studied.

nucl-th↗

arXiv2017

Interference Between Light and Heavy Neutrinos for $0νββ$ Decay in the Left-Right Symmetric Model

Neutrinoless double-beta decay is proposed as an important low energy phenomenon that could test beyond the Standard Model physics. There are several potentially competing beyond the Standard Model mechanisms that can induce the process. It thus becomes important to disentangle the different processes. In the present study we consider the interference effect between the light left-handed and heavy right-handed Majorana neutrino exchange mechanisms. The decay rate, and consequently, the phase-space factors for the interference term are derived, based on the left-right symmetric model. The numerical values for the interference phase-space factors for several nuclides are calculated, taking into consideration the relativistic Coulomb distortion of the electron wave function and finite-size of the nucleus. The variation of the interference effect with the Q-value of the process is studied.

hep-ph↗