Saeedeh Momtazi

Fatemeh Ziaeetabar, Reza Safabakhsh, Saeedeh Momtazi et al.

Nuanced understanding and the generation of detailed descriptive content for (bimanual) manipulation actions in videos is important for disciplines such as robotics, human-computer interaction, and video content analysis. This study describes a novel method, integrating graph based modeling with layered hierarchical attention mechanisms, resulting in higher precision and better comprehensiveness of video descriptions. To achieve this, we encode, first, the spatio-temporal inter dependencies between objects and actions with scene graphs and we combine this, in a second step, with a novel 3-level architecture creating a hierarchical attention mechanism using Graph Attention Networks (GATs). The 3-level GAT architecture allows recognizing local, but also global contextual elements. This way several descriptions with different semantic complexity can be generated in parallel for the same video clip, enhancing the discriminative accuracy of action recognition and action description. The performance of our approach is empirically tested using several 2D and 3D datasets. By comparing our method to the state of the art we consistently obtain better performance concerning accuracy, precision, and contextual relevance when evaluating action recognition as well as description generation. In a large set of ablation experiments we also assess the role of the different components of our model. With our multi-level approach the system obtains different semantic description depths, often observed in descriptions made by different people, too. Furthermore, better insight into bimanual hand-object interactions as achieved by our model may portend advancements in the field of robotics, enabling the emulation of intricate human actions with heightened precision.

Fatemeh Ziaeetabar, Reza Safabakhsh, Saeedeh Momtazi et al.

Automatic video description requires the generation of natural language statements about the actions, events, and objects in the video. An important human trait, when we describe a video, is that we are able to do this with variable levels of detail. Different from this, existing approaches for automatic video descriptions are mostly focused on single sentence generation at a fixed level of detail. Instead, here we address video description of manipulation actions where different levels of detail are required for being able to convey information about the hierarchical structure of these actions relevant also for modern approaches of robot learning. We propose one hybrid statistical and one end-to-end framework to address this problem. The hybrid method needs much less data for training, because it models statistically uncertainties within the video clips, while in the end-to-end method, which is more data-heavy, we are directly connecting the visual encoder to the language decoder without any intermediate (statistical) processing step. Both frameworks use LSTM stacks to allow for different levels of description granularity and videos can be described by simple single-sentences or complex multiple-sentence descriptions. In addition, quantitative results demonstrate that these methods produce more realistic descriptions than other competing approaches.

Niloofar Ranjbar, Saeedeh Momtazi, MohammadMehdi Homayounpour

Currently, there is a significant amount of research being conducted in the field of artificial intelligence to improve the explainability and interpretability of deep learning models. It is found that if end-users understand the reason for the production of some output, it is easier to trust the system. Recommender systems are one example of systems that great efforts have been conducted to make their output more explainable. One method for producing a more explainable output is using counterfactual reasoning, which involves altering minimal features to generate a counterfactual item that results in changing the output of the system. This process allows the identification of input features that have a significant impact on the desired output, leading to effective explanations. In this paper, we present a method for generating counterfactual explanations for both tabular and textual features. We evaluated the performance of our proposed method on three real-world datasets and demonstrated a +5\% improvement on finding effective features (based on model-based measures) compared to the baseline method.

Mohammad Hadi Goldani, Reza Safabakhsh, Saeedeh Momtazi

News consumption has significantly increased with the growing popularity and use of web-based forums and social media. This sets the stage for misinforming and confusing people. To help reduce the impact of misinformation on users' potential health-related decisions and other intents, it is desired to have machine learning models to detect and combat fake news automatically. This paper proposes a novel transformer-based model using Capsule neural Networks(CapsNet) called X-CapsNet. This model includes a CapsNet with dynamic routing algorithm paralyzed with a size-based classifier for detecting short and long fake news statements. We use two size-based classifiers, a Deep Convolutional Neural Network (DCNN) for detecting long fake news statements and a Multi-Layer Perceptron (MLP) for detecting short news statements. To resolve the problem of representing short news statements, we use indirect features of news created by concatenating the vector of news speaker profiles and a vector of polarity, sentiment, and counting words of news statements. For evaluating the proposed architecture, we use the Covid-19 and the Liar datasets. The results in terms of the F1-score for the Covid-19 dataset and accuracy for the Liar dataset show that models perform better than the state-of-the-art baselines.

Arya Hadizadeh Moghaddam, Saeedeh Momtazi

The advancement of social media contributes to the growing amount of content they share frequently. This framework provides a sophisticated place for people to report various real-life events. Detecting these events with the help of natural language processing has received researchers' attention, and various algorithms have been developed for this goal. In this paper, we propose a Semantic Modular Model (SMM) consisting of 5 different modules, namely Distributional Denoising Autoencoder, Incremental Clustering, Semantic Denoising, Defragmentation, and Ranking and Processing. The proposed model aims to (1) cluster various documents and ignore the documents that might not contribute to the identification of events, (2) identify more important and descriptive keywords. Compared to the state-of-the-art methods, the results show that the proposed model has a higher performance in identifying events with lower ranks and extracting keywords for more important events in three English Twitter datasets: FACup, SuperTuesday, and USElection. The proposed method outperformed the best reported results in the mean keyword-precision metric by 7.9\%.

Farhan Farsi, Farnaz Aghababaloo, Shahriar Shariati Motlagh et al.

As large language models (LLMs) become increasingly embedded in our daily lives, evaluating their quality and reliability across diverse contexts has become essential. While comprehensive benchmarks exist for assessing LLM performance in English, there remains a significant gap in evaluation resources for other languages. Moreover, because most LLMs are trained primarily on data rooted in European and American cultures, they often lack familiarity with non-Western cultural contexts. To address this limitation, our study focuses on the Persian language and Iranian culture. We introduce 19 new evaluation datasets specifically designed to assess LLMs on topics such as Iranian law, Persian grammar, Persian idioms, and university entrance exams. Using these datasets, we benchmarked 41 prominent LLMs, aiming to bridge the existing cultural and linguistic evaluation gap in the field.

Mohammad Ghiasvand Mohammadkhani, Niloofar Ranjbar, Saeedeh Momtazi

Generative approaches have significantly influenced Aspect-Based Sentiment Analysis (ABSA), garnering considerable attention. However, existing studies often predict target text components monolithically, neglecting the benefits of utilizing single elements for tuple prediction. In this paper, we introduce Element to Tuple Prompting (E2TP), employing a two-step architecture. The former step focuses on predicting single elements, while the latter step completes the process by mapping these predicted elements to their corresponding tuples. E2TP is inspired by human problem-solving, breaking down tasks into manageable parts, using the first step's output as a guide in the second step. Within this strategy, three types of paradigms, namely E2TP($diet$), E2TP($f_1$), and E2TP($f_2$), are designed to facilitate the training process. Beyond dataset-specific experiments, our paper addresses cross-domain scenarios, demonstrating the effectiveness and generalizability of the approach. By conducting a comprehensive analysis on various benchmarks, we show that E2TP achieves new state-of-the-art results in nearly all cases.

Mohammad Heydari Rad, Rezvan Afari, Saeedeh Momtazi

Multiple-choice questions (MCQs) are commonly used in educational testing, as they offer an efficient means of evaluating learners' knowledge. However, generating high-quality MCQs, particularly in low-resource languages such as Persian, remains a significant challenge. This paper introduces FarsiMCQGen, an innovative approach for generating Persian-language MCQs. Our methodology combines candidate generation, filtering, and ranking techniques to build a model that generates answer choices resembling those in real MCQs. We leverage advanced methods, including Transformers and knowledge graphs, integrated with rule-based approaches to craft credible distractors that challenge test-takers. Our work is based on data from Wikipedia, which includes general knowledge questions. Furthermore, this study introduces a novel Persian MCQ dataset comprising 10,289 questions. This dataset is evaluated by different state-of-the-art large language models (LLMs). Our results demonstrate the effectiveness of our model and the quality of the generated dataset, which has the potential to inspire further research on MCQs.

Samin Mahdipour Aghabagher, Saeedeh Momtazi

Dialogue State Tracking (DST) is an essential element of conversational AI with the objective of deeply understanding the conversation context and leading it toward answering user requests. Due to high demands for open-domain and multi-turn chatbots, the traditional rule-based DST is not efficient enough, since it cannot provide the required adaptability and coherence for human-like experiences in complex conversations. This study proposes a hybrid DST model that utilizes rule-based methods along with language models, including BERT for slot filling and intent detection, XGBoost for intent validation, GPT for DST, and online agents for real-time answer generation. This model is uniquely designed to be evaluated on a comprehensive Persian multi-turn dialogue dataset and demonstrated significantly improved accuracy and coherence over existing methods in Persian-based chatbots. The results demonstrate how effectively a hybrid approach may improve DST capabilities, paving the way for conversational AI systems that are more customized, adaptable, and human-like.

Mohammad Hosseini, Kimia Hosseini, Shayan Bali et al.

Hallucination is a persistent issue affecting all large language Models (LLMs), particularly within low-resource languages such as Persian. PerHalluEval (Persian Hallucination Evaluation) is the first dynamic hallucination evaluation benchmark tailored for the Persian language. Our benchmark leverages a three-stage LLM-driven pipeline, augmented with human validation, to generate plausible answers and summaries regarding QA and summarization tasks, focusing on detecting extrinsic and intrinsic hallucinations. Moreover, we used the log probabilities of generated tokens to select the most believable hallucinated instances. In addition, we engaged human annotators to highlight Persian-specific contexts in the QA dataset in order to evaluate LLMs' performance on content specifically related to Persian culture. Our evaluation of 12 LLMs, including open- and closed-source models using PerHalluEval, revealed that the models generally struggle in detecting hallucinated Persian text. We showed that providing external knowledge, i.e., the original document for the summarization task, could mitigate hallucination partially. Furthermore, there was no significant difference in terms of hallucination when comparing LLMs specifically trained for Persian with others.

Kasra Darvishi, Newsha Shahbodagh, Zahra Abbasiantaeb et al.

We present Persian Question Answering Dataset (PQuAD), a crowdsourced reading comprehension dataset on Persian Wikipedia articles. It includes 80,000 questions along with their answers, with 25% of the questions being adversarially unanswerable. We examine various properties of the dataset to show the diversity and the level of its difficulty as an MRC benchmark. By releasing this dataset, we aim to ease research on Persian reading comprehension and development of Persian question answering systems. Our experiments on different state-of-the-art pre-trained contextualized language models show 74.8% Exact Match (EM) and 87.6% F1-score that can be used as the baseline results for further research on Persian QA.

Parisa Abolfath Beygi Dezfouli, Saeedeh Momtazi, Mehdi Dehghan

Users' reviews contain valuable information which are not taken into account in most recommender systems. According to the latest studies in this field, using review texts could not only improve the performance of recommendation, but it can also alleviate the impact of data sparsity and help to tackle the cold start problem. In this paper, we present a neural recommender model which recommends items by leveraging user reviews. In order to predict user rating for each item, our proposed model, named MatchPyramid Recommender System (MPRS), represents each user and item with their corresponding review texts. Thus, the problem of recommendation is viewed as a text matching problem such that the matching score obtained from matching user and item texts could be considered as a good representative of their joint extent of similarity. To solve the text matching problem, inspired by MatchPyramid (Pang, 2016), we employed an interaction-based approach according to which a matching matrix is constructed given a pair of input texts. The matching matrix, which has the property of hierarchical matching patterns, is then fed into a Convolutional Neural Network (CNN) to compute the matching score for the given user-item pair. Our experiments on the small data categories of Amazon review dataset show that our proposed model gains from 1.76% to 21.72% relative improvement compared to DeepCoNN model, and from 0.83% to 3.15% relative improvement compared to TransNets model. Also, on two large categories, namely AZ-CSJ and AZ-Mov, our model achieves relative improvements of 8.08% and 7.56% compared to the DeepCoNN model, and relative improvements of 1.74% and 0.86% compared to the TransNets model, respectively.

Amin Abolghasemi, Saeedeh Momtazi

Knowledge graphs are widely used as a typical resource to provide answers to factoid questions. In simple question answering over knowledge graphs, relation extraction aims to predict the relation of a factoid question from a set of predefined relation types. Most recent methods take advantage of neural networks to match a question with all predefined relations. In this paper, we propose an instance-based method to capture the underlying relation of question and to this aim, we detect matching paraphrases of a new question which share the same relation, and their corresponding relation is selected as our prediction. The idea of our model roots in the fact that a relation can be expressed with various forms of questions while these forms share lexically or semantically similar terms and concepts. Our experiments on the SimpleQuestions dataset show that the proposed model achieves better accuracy compared to the state-of-the-art relation extraction models.

Zahra Abbasiantaeb, Saeedeh Momtazi

Text-based Question Answering (QA) is a challenging task which aims at finding short concrete answers for users' questions. This line of research has been widely studied with information retrieval techniques and has received increasing attention in recent years by considering deep neural network approaches. Deep learning approaches, which are the main focus of this paper, provide a powerful technique to learn multiple layers of representations and interaction between questions and texts. In this paper, we provide a comprehensive overview of different models proposed for the QA task, including both traditional information retrieval perspective, and more recent deep neural network perspective. We also introduce well-known datasets for the task and present available results from the literature to have a comparison between different techniques.

Mohammad Hadi Goldani, Saeedeh Momtazi, Reza Safabakhsh

Fake news is dramatically increased in social media in recent years. This has prompted the need for effective fake news detection algorithms. Capsule neural networks have been successful in computer vision and are receiving attention for use in Natural Language Processing (NLP). This paper aims to use capsule neural networks in the fake news detection task. We use different embedding models for news items of different lengths. Static word embedding is used for short news items, whereas non-static word embeddings that allow incremental up-training and updating in the training phase are used for medium length or large news statements. Moreover, we apply different levels of n-grams for feature extraction. Our proposed architectures are evaluated on two recent well-known datasets in the field, namely ISOT and LIAR. The results show encouraging performance, outperforming the state-of-the-art methods by 7.8% on ISOT and 3.1% on the validation set, and 1% on the test set of the LIAR dataset.