http://repository.iiitd.edu.in/xmlui/handle/123456789/1711 Year-2025 Thu, 03 Sep 2026 01:08:57 GMT 2026-09-03T01:08:57Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1996 Bridging the gap: developing robust and ethical AI systems Chhabra, Saheb; Vatsa, Mayank (Advisor); Singh, Richa (Advisor); Gupta, Gaurav (Advisor) Machine learning and deep learning algorithms have demonstrated exceptional performance across various computer vision tasks, including face detection, attribute prediction, object detection, image segmentation, and classification. The capabilities of deep neural networks (DNNs) have driven the development of numerous computer vision- based artificial intelligence (AI) systems, employed in diverse applications such as autonomous vehicles, COVID-19 detection in X-ray images, facial recognition for security and surveillance systems. Despite these advancements, the real-world applicability of such systems remains a critical challenge due to issues such as the unethical exploitation of AI technologies and failures of deep models on real-world datasets. One significant concern is the unethical exploitation of AI systems, particularly in contexts involving privacy violations and the generation of fake media. Instances such as unauthorized profiling for targeted advertising and the creation and dissemination of Deepfake content pose serious ethical and social risks. These activities not only infringe on individual privacy but also create widespread societal distress, potentially leading to psychological harm. Another challenge lies in the failure of deep models to generalize effectively to unseen real-world datasets. This is largely due to a lack of robustness and generalizability in model training, as well as data distribution shifts during testing. For example, a model trained on the Labeled Faces in the Wild (LFW) dataset containing over 13,000 unconstrained facial images collected from the web, may perform poorly when evaluated on the CelebA dataset, which consists of more than 200,000 celebrity face images annotated with 40 facial attributes. This performance drop occurs because of discrepancies in data characteristics. To address these challenges, this dissertation investigates the problems in the input space, the model space, and the feature space, identifying the input space as the root cause. The lack of a control mechanism to guide the attention of deep models toward meaningful image features underpins these issues. To resolve this, we propose novel solutions leveraging the dynamics of attention mechanisms in deep models. Two solutions are introduced to mitigate the unethical exploitation of AI systems: 1) Privacy Preservation: A framework that enables users to selectively anonymize facial identity and attribute information in images. This allows users to control the attention of the model, blocking it from processing specific attributes such as gender or age. 2) Deepfake Detection: An approach tailored to detect fake media across varying compression levels and algorithms used on social media platforms. This solution improves detection accuracy by training models to identify unseen artifacts in low-quality compressed media. The latter part of this dissertation addresses deep model failures on real-world datasets with two innovative solutions: 1) Data Shift Mitigation: A method to counteract data distribution shifts by learning uniform perturbations in the dataset. These perturbations improve model attention toward meaningful features, ensuring better performance in real-world scenarios. 2) Training Robustness: A strategy to diversify attention of the model during training, compelling it to learn overlooked features within the dataset. This approach enhances the robustness and generalizability of the model by expanding the range of features it considers. Overall, this dissertation presents novel mechanisms to address critical challenges in AI system reliability and ethical deployment by focusing on privacy preservation, Deep- fake detection, and generalizable learning. These solutions contribute significantly to the development of robust, generalizable, and ethical AI systems for real-world applications. Sat, 01 Feb 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1996 2025-02-01T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1783 Learning-based 3D reconstruction and synthesis problems Mathur, Aradhya Neeraj; Sharma, Ojaswa (Advisor) Sat, 20 Sep 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1783 2025-09-20T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1780 Towards enhanced conversational dynamics for effective virtual therapist-assistive counseling Srivastava, Aseem; Akhtar, Md. Shad (Advisor); Chakraborty, Tanmoy (Advisor) Mental health issues are now the leading global cause of disability, with conditions such as anxiety and depression escalating, particularly following the COVID-19 pandemic. However, traditional mental health support methods remain heavily constrained by a severe shortage of trained professionals, making it difficult to meet the increasing demand for mental health support systems. In response, we propose Virtual Mental Health Assistants (VMHAs) as a scalable and accessible alternative, offering instant, anonymous, and stigma-free method of support. Despite their potential, existing automated counseling systems are limited by rigid, scripted dialogues and fail to replicate the nuanced adaptability and therapeutic depth of human therapists. To address this, we focus on dialogue systems on modular levels, spanning understanding, summarization, generation, and evaluation of therapeutic conversations. This thesis studies the counseling interaction pipeline, refining its core components to enhance the efficiency and effectiveness of professionals. The goal is to improve dialogue understanding, enabling VMHAs to interpret users’ implicit psychological intents through directive recognition. To maintain coherence and continuity across conversations, we incorporate domain knowledge infusion into counseling summarization, allowing the system to retain relevant memory. Additionally, we advance dialogue generation by integrating clinically informed, emotionally adaptive response models, surpassing traditional rule-based and purely generative approaches to ensure more human-like and therapeutic interactions. Next, we propose a dialogue evaluation framework that centers on therapeutic bond assessment via trust modeling. Recognizing that mental health support extends beyond one-on-one counseling, we further analyze peer interactions in online communities and group therapy, positioning AI as a facilitator of collective support environments. Through rigorous experimentation, user studies, and exhaustive analysis, this thesis establishes a carefully designed, context-aware, and psychologically informed understanding of counseling interactions. Rather than approaching VMHAs as standalone interventions, this research emphasizes their role as an augmentative approach, proposed to enhance the efficiency of mental health professionals. We conduct each phase of this study in close collaboration with domain experts, ensuring that the proposed methodologies are novel and practically viable in real-world settings. Moreover, the findings presented here extend beyond novel methodological contributions, positioning this thesis’ findings not just as novel solutions, but as supportive alternatives that alleviate the burden on mental health professionals. Fri, 10 Oct 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1780 2025-10-10T00:00:00Z http://repository.iiitd.edu.in/xmlui/handle/123456789/1778 A framework for non-factoid question answering in Indic languages Mishra, Ritwik; Shah, Rajiv Ratn (Advisor); Kumaraguru, Ponnurangam (Advisor) We address the development of explainable Question Answering (QA) systems for Indic languages, focusing on the unique challenges posed by resource scarcity and the complexities of multilingual processing. The research begins by categorizing QA systems based on context, domain, conversational requirements, and answer types, emphasizing the importance of text-based QA for cognitive development. A comprehensive literature review highlights advances in factoid and non-factoid QA, the rise of Transformer-based models, and the critical role of retrieval mechanisms for handling extended contexts. Our work also identifies significant gaps in resources for Indic languages, particularly for non-factoid QA, and underscores the necessity for efficient, explainable, and retrieval-augmented models. To address the lack of structured knowledge extraction tools for low-resource languages, the thesis introduces IndIE, an Open Information Extraction (OIE) system designed for Hindi. IndIE employs a multilingual pretrained transformer, fine-tuned on chunk-annotated data from English and five Indic languages, to generate triples from unstructured sentences. In sequence labeling tasks (like chunking), it was found that the mean of subword token embeddings is more beneficial than other approaches. The system leverages chunk tagging and Merged-Phrase Dependency Trees, achieving a 0.51 F1-score on a benchmark of 112 Hindi sentences and producing more granular triples than existing multilingual approaches. The underlying methodology demonstrates potential for extension to Urdu, Tamil, and Telugu, given the generalizability of the chunker and the language-agnostic nature of the triple extraction rules. Recognizing the challenge of resolving references to the same entity across text, the thesis present TransMuCoRes, a multilingual coreference resolution dataset spanning 31 South Asian languages. Using automated translation and word alignment, TransMuCoRes fills a critical resource gap for coreference tasks in these languages. Two coreference models, trained on a combination of TransMuCoRes and manually annotated Hindi data, achieve LEA F1 and CoNLL F1 scores of 64 and 68, respectively, on a Hindi test set. The work also critiques current evaluation metrics, advocating for improved measures to handle split antecedents. Building on these foundational tools, the thesis introduces MuNfQuAD, a multilingual non-factoid QA dataset comprising over 578K question-answer pairs across 38 languages, including numerous low-resource languages. Questions are derived from interrogative sub-headings in BBC news articles, with corresponding paragraphs serving as silver-standard answers. Manual annotation of 790 pairs reveals that 98% of questions are answerable using the provided context. An Answer Paragraph Selection (APS) model, fine- tuned on this dataset, achieves 80% accuracy and 72% macro F1 on the test set, and 72% accuracy and 66% macro F1 on the golden set, outperforming baseline methods and demon- strating effective context reduction. The thesis further investigates explainability in QA and related tasks. Through experiments on the HateXplain benchmark, it compares three post-hoc interpretability methods for transformer-based encoders in hate speech detection. Notably, Layerwise Relevance Propagation (LRP) underperforms, sometimes even less informative than random rationale generation, due to its tendency to focus on initial tokens. This finding highlights the limitations of LRP for explaining fine-tuned transformer predictions. To enhance QA performance for long contexts, especially in Indic languages, the thesis explores various context-shortening strategies based on OIE, coreference resolution, and APS. Experiments with three popular Large Language Models (LLMs) on Hindi, Tamil, Telugu, and Urdu show that these techniques improve semantic scores by an average of 4% and token-level scores by 47% without fine-tuning, and by 2% with fine-tuning, while also reducing computational demands. Explain ability analyses using LIME and SHAP indicate that APS-selected para- graphs concentrate model attention on relevant tokens. However, the study notes persistent challenges for LLMs in non-factoid QA requiring reasoning, and finds that verbalizing OIE triples does not further enhance performance. As a retrospective epilogue of the thesis, we also present a Hindi chatbot for maternal and child health queries. Using a curated FAQ database and an ensemble of rule-based, embedding-based, and paraphrasing classifiers, the system covers 80% of user queries and retrieves at least one relevant answer in the top three suggestions for 70% of cases. Collectively, this work advances the state of explainable QA for Indic languages by developing novel resources, tools, and evaluation frameworks, and by demonstrating the effectiveness of context-shortening and interpretability techniques in low-resource, multilingual settings. Future work in QA systems for Indic languages includes expanding benchmarks like Hindi- BenchIE to other low-resource languages for standardized evaluation of triple extraction methods, thus advancing multilingual OIE. The release of TransMuCoRes checkpoints offers a baseline for multilingual coreference resolution research. Using APS models as reward models for LLM alignment may improve answer accuracy for complex queries. Additional directions involve deploying chatbots in real-world settings, refining OIE and coreference models, expanding multilingual QA datasets, and enhancing explainability. Evaluating systems on longer contexts and integrating advanced alignment strategies will foster robust, transparent QA frameworks for Indic languages. Thu, 01 May 2025 00:00:00 GMT http://repository.iiitd.edu.in/xmlui/handle/123456789/1778 2025-05-01T00:00:00Z