arXiv subjects

Behnam Bahrak

Publications and source records attributed to Behnam Bahrak.

At least 19 recordsLinked to original sources

arXiv2026

PERCEPT: A Corpus for POS Tagging and Analysis of Persian-English Code-Mixing

Social media has become a major venue for multilingual communication, where users frequently mix multiple languages within a single utterance. Although code-mixed corpora have been developed for several language pairs, Persian-English code-mixing remains relatively underexplored. Existing Persian resources lack Universal Dependencies (UD) part-of-speech (POS) annotations for code-mixed words, limiting both linguistic analyses and the development of syntax-aware NLP models. To address this gap, we introduce PERCEPT, the first publicly available large-scale Persian-English code-mixed corpus annotated with Universal Dependencies POS tags for code-mixed words. The dataset comprises 6,800 posts collected from X, Instagram, and Digikala. We further present an LLM-assisted annotation framework that automatically assigns POS tags and document-level topics. Human evaluation demonstrates high agreement between the automatically generated annotations and gold annotations, confirming the reliability of the annotations. Using PERCEPT, we conduct the first comprehensive linguistic analysis of Persian-English code-mixing across multiple social media platforms. Our analyses reveal that nouns are the predominant category for code-mixed words, while the distributions of other POS categories vary across platforms. We further find that the positional distribution of code-mixed words is remarkably consistent across platforms, whereas the triggering effect is substantially more pronounced in Digikala. PERCEPT is publicly available at https://github.com/kalhorghazal/PERCEPT.

cs.CL↗

arXiv2026

PsychoAgent: An Affect-Sensitive Cognitive Architecture for Conflict-Aware Memory in LLM Agents

Human-like cognition does not select past experience by topical similarity alone: affective significance and unresolved conflict also shape what becomes accessible. We present PsychoAgent, a cognitive architecture for LLM agents that separates factual and affective memory and integrates both through a conflict-aware executive controller. Affective memories are first filtered by semantic relevance and then re-ranked by salience, preserving topical fit while allowing emotionally important traces to enter the prompt. Across three controlled conflict scenarios, the full architecture retrieved more conflict-critical memories than semantic-affective and single-memory RAG baselines (0.933 vs. 0.500 and 0.667), with a small semantic-similarity cost. Five blinded raters evaluated 27 outputs. After within-rater standardization, the full architecture had the highest overall mean (+0.22 SD), but corrected pairwise differences were not significant. A three-day illustrative trace further shows persistent affect, offline memory recombination, and selective memory reweighting. The findings support affect-sensitive retrieval as an inspectable mechanism for modeling human-like conflict effects in LLM agents.

cs.AI↗

arXiv2026

Generating Bearing Vibration Signals at User-Specified Fault Probabilities Using PR-GAN and Counterfactual Methods

In bearing vibration datasets, most samples receive predicted fault probabilities close to 0 or 1, while samples with intermediate (gray-zone) probabilities are rare. Such borderline samples are important because they reflect conditions in which maintenance decisions may require additional inspection or a conservative response and are useful for studying decision boundaries. To address this scarcity, this paper proposes and compares two approaches that generate vibration signals whose predicted fault probability matches a target probability of 0.25, 0.50, or 0.75. We use the average output of a heterogeneous ensemble classifier with different architectures and random initializations as a fixed, gradient-accessible probability oracle. The first, training-based approach, Probability-Regularized Generative Adversarial Network (PR-GAN), extends Wasserstein Generative Adversarial Network with Gradient Penalty (WGAN-GP) and edits a real signal through a residual generator while pushing the classifier output toward the target probability. The second is a training-free, per-sample Wachter-style counterfactual (CF) procedure that directly optimizes each input signal to reach the target probability while remaining close to the source signal. We evaluate both methods on the Case Western Reserve University (CWRU) and Paderborn bearing datasets using mean absolute target-probability error, time-domain total variation, and frequency-domain log power spectral density (log-PSD) differences. Across all settings, CF reaches the target with a mean absolute probability error of 0.005-0.008 and a within-tolerance success rate of 1.000 on retained samples, whereas PR-GAN's mean error is 0.046-0.059 with success rates between 0.501 and 0.680. CF therefore steers the probability more reliably and requires smaller average L1 changes, whereas PR-GAN has a lower reported runtime in most settings.

cs.LG↗

arXiv2026

Beyond Access: Guided LLM Scaffolding for Independent Learning in Undergraduate Statistics

Large language models (LLMs) are increasingly entering students' learning practices, but their educational value may depend on whether they are used to support reasoning or to complete tasks without engaging in the underlying reasoning. This study examines guided LLM use in an undergraduate Probability and Statistics course, focusing on the distinction between assigned LLM access and the quality of students' actual interaction with the model. In a four-week quasi-experimental summer program, students were organized into three balanced conditions: no LLM access, unrestricted LLM access, and guided LLM access. The guided condition used the same LLM platform as the unrestricted condition, but students received explicit training and rules intended to promote reasoning-focused help-seeking, stepwise hints, verification, and ethical use. All quizzes and the delayed final exam were completed without LLM or external assistance, allowing us to separate AI-supported practice performance from independent learning. Results show that guided use was associated with a clearer learning-oriented interaction pattern than unrestricted access, especially in prioritizing reasoning over final answers and requesting stepwise support. In behavior-defined analyses, Guided-LLM students showed a promising pattern of stronger no-help quiz performance, while practice scores showed no consistent Guided-LLM advantage. Available time measures did not support a simple duration-based explanation, and self-assessment calibration suggested better alignment between perceived and demonstrated understanding in Guided-LLM. These findings suggest that access alone may not reliably distinguish independent performance; instead, the quality of interaction and reasoning-focused scaffolds warrant further study.

cs.CY↗

arXiv2026

TopRank-Based Delivery Rate Optimization for Coded Caching under Non-Uniform Demands

We study the problem of coded caching with nonuniform file popularity under the setting where the popularity distribution is initially unknown. By reframing the problem, we propose a method inspired by an algorithm from the recommender-systems literature and multi-armed bandits. Unlike prior approaches, which focus on accurately estimating file popularities, our method ranks files relative to one another and partitions them into groups. This perspective is more consistent with the structure of prior approaches as well, since earlier methods also divided files into popular and non-popular groups after estimating their popularities. The proposed approach relies on differences in request counts between files as the basis for ranking, and under many conditions it outperforms the previous algorithm. In particular, we obtain significantly improved performance in scenarios where the number of users in the network is small, the cache storage capacity is limited, or the learning process of the true popularity of files based on observations is contaminated by exploratory or synthetic requests that do not match the true popularity distribution. In these cases, our policy achieves markedly better performance and attains sublinear regret.

cs.CR↗

arXiv2026

KNIGHT: Knowledge Graph-Driven Multiple-Choice Question Generation with Adaptive Hardness Calibration

With the rise of large language models (LLMs), they have become instrumental in applications such as Retrieval-Augmented Generation (RAG). Yet evaluating these systems remains bottlenecked by the time and cost of building specialized assessment datasets. We introduce KNIGHT, an LLM-based, knowledge-graph-driven framework for generating multiple-choice question (MCQ) datasets from external sources. KNIGHT constructs a topic-specific knowledge graph, a structured and parsimonious summary of entities and relations, that can be reused to generate instructor-controlled difficulty levels, including multi-hop questions, without repeatedly re-feeding the full source text. This knowledge graph acts as a compressed, reusable state, making question generation a cheap read over the graph. We instantiate KNIGHT on Wikipedia/Wikidata while keeping the framework domain- and ontology-agnostic. As a case study, KNIGHT produces six MCQ datasets in History, Biology, and Mathematics. We evaluate quality on five criteria: fluency, unambiguity (single correct answer), topic relevance, option uniqueness, and answerability given the provided sources (as a proxy for hallucination). Results show that KNIGHT enables token- and cost-efficient generation from a reusable graph representation, achieves high quality across these criteria, and yields model rankings aligned with MMLU-style benchmarks, while supporting topic-specific and difficulty-controlled evaluation.

cs.CL↗

arXiv2026

The Sufficiency-Conciseness Trade-off in LLM Self-Explanation from an Information Bottleneck Perspective

Large Language Models increasingly rely on self-explanations, such as chain of thought reasoning, to improve performance on multi step question answering. While these explanations enhance accuracy, they are often verbose and costly to generate, raising the question of how much explanation is truly necessary. In this paper, we examine the trade-off between sufficiency, defined as the ability of an explanation to justify the correct answer, and conciseness, defined as the reduction in explanation length. Building on the information bottleneck principle, we conceptualize explanations as compressed representations that retain only the information essential for producing correct answers.To operationalize this view, we introduce an evaluation pipeline that constrains explanation length and assesses sufficiency using multiple language models on the ARC Challenge dataset. To broaden the scope, we conduct experiments in both English, using the original dataset, and Persian, as a resource-limited language through translation. Our experiments show that more concise explanations often remain sufficient, preserving accuracy while substantially reducing explanation length, whereas excessive compression leads to performance degradation.

cs.CL↗

arXiv2026

A Hybrid TGN-SEAL Model for Dynamic Graph Link Prediction

Predicting links in sparse, continuously evolving networks is a central challenge in network science. Conventional heuristic methods and deep learning models, including Graph Neural Networks (GNNs), are typically designed for static graphs and thus struggle to capture temporal dependencies. Snapshot-based techniques partially address this issue but often encounter data sparsity and class imbalance, particularly in networks with transient interactions such as telecommunication call detail records (CDRs). Temporal Graph Networks (TGNs) model dynamic graphs by updating node embeddings over time; however, their predictive accuracy under sparse conditions remains limited. In this study, we improve the TGN framework by extracting enclosing subgraphs around candidate links, enabling the model to jointly learn structural and temporal information. Experiments on a sparse CDR, email, message dataset show that our approach increases average precision by at least 2% over standard TGNs, demonstrating the advantages of integrating local topology for robust link prediction in dynamic networks.

cs.SI↗

arXiv2026

MasalBench: A Benchmark for Contextual and Cross-Cultural Understanding of Persian Proverbs in LLMs

In recent years, multilingual Large Language Models (LLMs) have become an inseparable part of daily life, making it crucial for them to master the rules of conversational language in order to communicate effectively with users. While previous work has evaluated LLMs' understanding of figurative language in high-resource languages, their performance in low-resource languages remains underexplored. In this paper, we introduce MasalBench, a comprehensive benchmark for assessing LLMs' contextual and cross-cultural understanding of Persian proverbs, which are a key component of conversation in this low-resource language. We evaluate eight state-of-the-art LLMs on MasalBench and find that they perform well in identifying Persian proverbs in context, achieving accuracies above 0.90. However, their performance drops considerably when tasked with identifying equivalent English proverbs, with the best model achieving 0.79 accuracy. Our findings highlight the limitations of current LLMs in cultural knowledge and analogical reasoning, and they provide a framework for assessing cross-cultural understanding in other low-resource languages. MasalBench is available at https://github.com/kalhorghazal/MasalBench.

cs.CL↗

arXiv2026

ArtCognition: A Multimodal AI Framework for Affective State Sensing from Visual and Kinematic Drawing Cues

The objective assessment of human affective and psychological states presents a significant challenge, particularly through non-verbal channels. This paper introduces digital drawing as a rich and underexplored modality for affective sensing. We present a novel multimodal framework, named ArtCognition, for the automated analysis of the House-Tree-Person (HTP) test, a widely used psychological instrument. ArtCognition uniquely fuses two distinct data streams: static visual features from the final artwork, captured by computer vision models, and dynamic behavioral kinematic cues derived from the drawing process itself, such as stroke speed, pauses, and smoothness. To bridge the gap between low-level features and high-level psychological interpretation, we employ a Retrieval-Augmented Generation (RAG) architecture. This grounds the analysis in established psychological knowledge, enhancing explainability and reducing the potential for model hallucination. Our results demonstrate that the fusion of visual and behavioral kinematic cues provides a more nuanced assessment than either modality alone. We show significant correlations between the extracted multimodal features and standardized psychological metrics, validating the framework's potential as a scalable tool to support clinicians. This work contributes a new methodology for non-intrusive affective state assessment and opens new avenues for technology-assisted mental healthcare.

cs.LG↗

arXiv2025

Can LLMs Faithfully Explain Themselves in Low-Resource Languages? A Case Study on Emotion Detection in Persian

Large language models (LLMs) are increasingly used to generate self-explanations alongside their predictions, a practice that raises concerns about the faithfulness of these explanations, especially in low-resource languages. This study evaluates the faithfulness of LLM-generated explanations in the context of emotion classification in Persian, a low-resource language, by comparing the influential words identified by the model against those identified by human annotators. We assess faithfulness using confidence scores derived from token-level log-probabilities. Two prompting strategies, differing in the order of explanation and prediction (Predict-then-Explain and Explain-then-Predict), are tested for their impact on explanation faithfulness. Our results reveal that while LLMs achieve strong classification performance, their generated explanations often diverge from faithful reasoning, showing greater agreement with each other than with human judgments. These results highlight the limitations of current explanation methods and metrics, emphasizing the need for more robust approaches to ensure LLM reliability in multilingual and low-resource contexts.

cs.CL↗

arXiv2025

Probing Gender Bias in Multilingual LLMs: A Case Study of Stereotypes in Persian

Multilingual Large Language Models (LLMs) are increasingly used worldwide, making it essential to ensure they are free from gender bias to prevent representational harm. While prior studies have examined such biases in high-resource languages, low-resource languages remain understudied. In this paper, we propose a template-based probing methodology, validated against real-world data, to uncover gender stereotypes in LLMs. As part of this framework, we introduce the Domain-Specific Gender Skew Index (DS-GSI), a metric that quantifies deviations from gender parity. We evaluate four prominent models, GPT-4o mini, DeepSeek R1, Gemini 2.0 Flash, and Qwen QwQ 32B, across four semantic domains, focusing on Persian, a low-resource language with distinct linguistic features. Our results show that all models exhibit gender stereotypes, with greater disparities in Persian than in English across all domains. Among these, sports reflect the most rigid gender biases. This study underscores the need for inclusive NLP practices and provides a framework for assessing bias in other low-resource languages.

cs.CL↗

arXiv2024

Understanding Trends, Patterns, and Dynamics in Global Company Acquisitions: A Network Perspective

Studying acquisitions offers invaluable insights into startup trends, aiding informed investment decisions for businesses. However, the scarcity of studies in this domain prompts our focus on shedding light in this area. Employing Crunchbase data, our study delves into the global network of company acquisitions using diverse network analysis techniques. Our findings unveil an acquisition network characterized by a primarily sparse structure comprising localized dense connections. We reveal a prevalent tendency among organizations to acquire companies within their own country and industry, as well as those within the same age bracket. Furthermore, we show that the country, region, city, and category of the companies can affect the formation of acquisition relationships between them. Our temporal analysis indicates a growth in the number of weakly connected components of the network over time, accompanied by a trend toward a sparser network. Through centrality metrics computation in the cross-city acquisition network, we identify New York, London, and San Francisco as pivotal and central hubs in the global economic landscape. Finally, we show that the United States, United Kingdom, and Germany are predominant countries in international acquisitions. The insights from our research assist policymakers in crafting better regulations to foster global economic growth, and aid businesses in deciding which startups to acquire and which markets to target for expansion.

cs.SI↗

arXiv2023

Analysis of Research Trends in Computer Science: A Network Approach

Nowadays, computer science (CS) has emerged as a dominant force in numerous research areas both within and beyond its own discipline. However, despite its significant impact on scholarly space, only a limited number of studies have been conducted to analyze the research trends and relationships within computer science. In this study, we collected information on fields and subfields from over 2,000 research articles published in the 2022 proceedings of the top Association for Computing Machinery (ACM) conferences spanning various research fields. Through a network approach, we investigated the interconnections between CS fields and subfields to evaluate their interdisciplinarity and multidisciplinarity. Our findings indicate that computing methodologies and privacy and security stand out as the most interdisciplinary fields, while human-centered computing exhibits the highest frequency among the papers. Furthermore, we discovered that machine learning emerges as the most interdisciplinary and multidisciplinary subfield within computer science. These results offer valuable insights for universities seeking to foster interdisciplinary research opportunities for their students.

cs.SI↗

arXiv2023

Forecasting Success of Computer Science Professors and Students Based on Their Academic and Personal Backgrounds

After completing their undergraduate studies, many computer science (CS) students apply for competitive graduate programs in North America. Their long-term goal is often to be hired by one of the big five tech companies or to become a faculty member. Therefore, being aware of the role of admission criteria may help them choose the best path towards their goals. In this paper, we analyze the influence of students' previous universities on their chances of being accepted to prestigious North American universities and returning to academia as professors in the future. Our findings demonstrate that the ranking of their prior universities is a significant factor in achieving their goals. We then illustrate that there is a bias in the undergraduate institutions of students admitted to the top 25 computer science programs. Finally, we employ machine learning models to forecast the success of professors at these universities. We achieved an RMSE of 7.85 for this prediction task.

cs.CY↗

arXiv2023

Mixing Solutions in Bitcoin and Ethereum Ecosystems: A Review and Tutorial

This manuscript presents an exhaustive review of blockchain-based mixing services, aiming to fill the existing gap between academic innovations and real-world implementations. Starting with an identification of the core functionalities and techniques employed by mixing services, the paper delves into detailed explanations of these operational mechanisms. It further outlines an evaluation framework tailored for a rigorous assessment, highlighting the key vulnerabilities and strengths of various solutions. In addition, the study identifies potential attack vectors that compromise these services. The paper explores the dual nature of mixing services, while they contribute to the preservation of privacy, a cornerstone of blockchain technologies, they can also facilitate illicit activities. By addressing key research questions, this study not only offers a comprehensive overview of the current state of mixing services but also sets the stage for future academic discourse in this evolving field.

cs.CR↗

arXiv2023

A Social Network Approach to Analyzing Token Properties and Abnormal Events in Decentralized Exchanges

The properties of tokens within the Ethereum blockchain, such as their current prices, trade volumes, and potential future values, have been the subjects of numerous studies. Employing social networks and graphs, as powerful tools for modeling connections within groups or communities would provide valuable guidance for analyzing these properties. This study mainly focuses on creating and examining networks related to two major decentralized exchanges including Uniswap Version 2 (UniswapV2) and SushiSwap. We have discovered that the distribution of nodes' degrees follows a power law that makes them scale-free networks, in addition, the centrality of tokens in exchange graphs provides valuable insights into their price and significance in cryptocurrency markets. These measures of centrality can be used to detect anomalies in cryptocurrency markets and prices. Notably, these networks exhibit remarkably similar structures, hinting at exciting research opportunities for modeling such networks.

cs.SI↗

arXiv2023

A Large-Scale Analysis of Persian Tweets Regarding Covid-19 Vaccination

The Covid-19 pandemic had an enormous effect on our lives, especially on people's interactions. By introducing Covid-19 vaccines, both positive and negative opinions were raised over the subject of taking vaccines or not. In this paper, using data gathered from Twitter, including tweets and user profiles, we offer a comprehensive analysis of public opinion in Iran about the Coronavirus vaccines. For this purpose, we applied a search query technique combined with a topic modeling approach to extract vaccine-related tweets. We utilized transformer-based models to classify the content of the tweets and extract themes revolving around vaccination. We also conducted an emotion analysis to evaluate the public happiness and anger around this topic. Our results demonstrate that Covid-19 vaccination has attracted considerable attention from different angles, such as governmental issues, safety or hesitancy, and side effects. Moreover, Coronavirus-relevant phenomena like public vaccination and the rate of infection deeply impacted public emotional status and users' interactions.

cs.CL↗