Abstract

r exaggerate content to drive user engagement. While they may boost clicks, such headlines and articles compromise reader trust and contribute to the spread of misinformation. Tackling clickbait thus represents not just a technical challenge but also a crucial step toward improving the quality and transparency of online information. To address this, recent research has introduced the task of clickbait spoiling, where we attempt to ”spoil” the headline by generating the key information the article withholds. Our work contributes to this growing area by firstly engaging with the clickbait spoiling task in English using modern Large Language Models (LLMs) and secondly, in Hindi, where we extend this task into a vastly underexplored linguistic space. The spoiling task is especially meaningful because it does not attempt to censor or suppress content instead it just simply makes it more honest, giving users the context they need upfront to make informed choices about what they click. From the evaluations on the English dataset it is revealed that LLMs, particularly when prompted with a few-shot setup, can be highly effective at generating relevant and informative spoilers. This demonstrates the potential of prompting-based systems to perform sophisticated reasoning and summarization tasks, even in cases where training data may be limited. Building on these insights, we make a significant contribution by shifting focus to Hindi, a language that has been critically underrepresented in this field of NLP research despite being spoken by hundreds of millions. We curate the first-of-its-kind Hindi clickbait dataset, using automatically generated annotations from multiple powerful LLMs. This dataset will not only address a huge gap in multilingual clickbait research but also lay a foundation for future work in Hindi misinformation and media literacy. We also observed notable inconsistencies in the behavior of certain large language models. Hermes-3, in particular, exhibited significant divergence in both its clickbait classification judgments and its assigned ratings. It was unable to understand explicit instructions and hence in light of this unreliability, we made a principled decision to take it off the annotator lineup. This screams for quality control whenever using LLMs for such tasks. We then extended the spoiling task to Hindi using this dataset by introducing a selection process based on agreement across models, combined with manual curation, resulting in a dataset that captures the diverse forms of Hindi clickbait. We also classify spoiler types and evaluate LLM performance in zero-shot and few-shot settings. Our findings reveal clear patterns: model performance varies widely, prompting strategies significantly affect outcomes, and certain models (like Deepseek-R1) demonstrate robust generalization even in zero-shot settings.

Abstract

Telugu, a Dravidian language spoken by approximately 95.7 million people, faces significant barriers in digital ecosystems due to its agglutinative morphology and limited computational and informational resources. This thesis addresses these challenges through two transformative contributions: a contextaware morphological analyzer and the generation of 25,000 high-quality Telugu Wikipedia articles to enhance knowledge accessibility. To tackle the scarcity of annotated datasets, a novel 10,000-sentence Telugu corpus was developed, meticulously annotated with morphological features such as lexical category, gender, number, person, case, tense-aspect-mood (TAM), and clause markers. This dataset, capturing complex forms like adav ¯ allalMdark ¯ ¯ı (‘to all women’) and ambiguities like pagalu (‘day’ vs. ‘to break’), powers a Transformer-based analyzer. Fine-tuned on BERT-Te, a monolingual model pretrained on 80,15,588 Telugu sentences, the analyzer achieves F1 scores of 0.778 for gender and 0.602 for lexical category, outperforming multilingual models (mBERT, IndicBERT, XLM-R) by 20-25%. This enables precise parsing for applications like machine translation and sentiment analysis, addressing a critical gap in Telugu NLP infrastructure. Concurrently, a scalable pipeline leveraging Jinja2 templates and structured data from IMDb, USDA, IUCN, and Smithsonian generated 25,000 Wikipedia articles across movies (8,929), plants (2,140), animals (4,928), and volcanoes (8,676). These articles, incorporating infoboxes, tables, and citations, achieve a BERTScore of 0.79 for semantic similarity and a Fleiss’ Kappa of 0.82 for human-rated quality, adhering to Wikipedia’s five pillars (neutrality, verifiability, no original research, free content, civility). Expanding Telugu Wikipedia’s 95,599-article base by over 25%, they empower native speakers with accessible knowledge in entertainment, science, and geography. Both contributions are opensourced at https://github.com/parameshkrishnaa/Telugu-Morph-Dataset and https://tewiki.iiit.ac.in, ensuring reproducibility and adaptability to other low-resource languages. Despite challenges like sandhi splitting (8% manual corrections) and transliteration errors (e.g., “The” as Te, 8% manual fixes), the ˆ research establishes robust benchmarks—F1 scores, BERTScore, and Fleiss’ Kappa—for Telugu NLP. By enhancing computational tools and public knowledge access, this work elevates Telugu’s digital presence, fosters linguistic equity, and provides a scalable model for global low-resource language advancement. Future directions include expanding datasets, automating preprocessing, and deploying articles on Wikipedia to further bridge the digital divide.

Abstract

Examining limitations is a crucial step in the scholarly research review process, revealing aspects where a study might lack decisiveness or require enhancement. This practice aids readers in considering broader implications for further research and helps in identifying potential areas for improvement and future investigation. In this thesis, we introduce a novel and challenging task of Suggestive Limitation Generation (SLG) for research papers. To facilitate this task, we compile a comprehensive dataset calling it LimGen, which encompasses 4068 research papers along with their associated limitations from the ACL anthology. The LimGen dataset represents a significant resource for the academic community, providing a structured and extensive collection of limitations that can be used to train and evaluate models aimed at generating meaningful and constructive limitations. We investigate several approaches to harness large language models (LLMs) for producing suggestive limitations. This involves a thorough examination of related challenges, practical insights, and potential opportunities. Our research delves into various techniques and methodologies, assessing their effectiveness and exploring ways to enhance the generation of suggestive limitations. The challenges addressed in this thesis include the inherent difficulty of understanding complex research content, the need for generating limitations that are both relevant and insightful, and the integration of domain-specific knowledge to improve the quality of the generated limitations. We also discuss the practical implications of our findings, providing valuable insights for researchers and practitioners looking to leverage LLMs for this purpose. We experimented with various setups, including using summarization-specific pre-trained models like BART and PEGASUS, and generative models such as T5, Cerebras-GPT, and Llama 2. To address the length constraint of input papers, we employed the Dense Passage Retrieval (DPR) approach, which processes relevant passages to maintain context. Additionally, we implemented a chain modeling technique to refine the generated limitations for better coherence and relevance. A comprehensive evaluation of the proposed models was conducted using automatic (ROUGE-based), human (manual expert-based), and model assistive (LLM-based) approaches. The results indicate that models trained specifically for summarization do not effectively generate insightful limitations, highlighting the complexity of the SLG task compared to traditional summarization tasks. Our contributions include the development of the LimGen dataset, the proposal of several LLM-based schemes for SLG, and a thorough evaluation of these methods. The LimGen dataset and accompanying code are made available to the research community to foster further exploration and development in this area. Our work aims to contribute to the advancement of automated tools for research evaluation, ultimately aiding in the improvement of research quality and the acceleration of scientific discovery

Abstract

In recent years, the widespread adoption of the internet and the growth in user-base of various social media platforms have led to a notable increase in the proliferation of extreme speech online. While traditional Natural Language Processing (NLP) models have demonstrated proficiency in distinguishing between neutral text and non-neutral text (i.e. extreme speech), categorizing the diverse types of extreme speech presents significant challenges. The task of extreme speech classification is particularly nuanced, as it requires a deep understanding of socio-cultural contexts to accurately interpret the intent of the language used by the speaker. Even human annotators often disagree on the appropriate classification of such content, emphasizing the complex and subjective nature of this task. The use of human moderators also presents a scaling issue, necessitating the need for automated systems for extreme speech classification. The recent launch of ChatGPT has drawn global attention to the potential applications of Large Language Models (LLMs) across a diverse array of tasks. Due to their training on vast and diverse corpora, coupled with their demonstrated ability to capture and encode contextual information effectively, LLMs emerge as highly promising tools for tackling this specific task of extreme speech classification. Their capacity to generalize across linguistic patterns and contextual nuances makes them particularly well-suited for applications requiring sophisticated language understanding. LLMs have the capability to understand complex language and context, making them useful tools for identifying and controlling extreme speech on online platforms. In this thesis, we leverage the Indian subset of the extreme speech dataset (XTREMESPEECH) introduced by Maronikolakis et al. [63] to evaluate zero-shot classification, finetuning, preference optimization, and ensemble methods. We address two primary research questions: (1) How do LLMs perform on multilingual extreme speech datasets, particularly within India’s linguistically and culturally diverse context? (2) To what extent can open-source LLMs serve as viable alternatives to proprietary GPT models, and does fine-tuning bridge any performance gap? Our findings indicate that while pre-trained LLMs demonstrate moderate efficacy in detecting extreme speech in a zero-shot setting, their performance is substantially enhanced through fine-tuning with domain-specific data. This improvement underscores the adaptability of these models in capturing linguistic subtleties and contextual nuances. Furthermore, we note that while GPT-based models exhibit superior performance compared to open-source Llama models in a zero-shot context, the performance gap vanishes after fine-tuning. This suggests that open-source LLMs, when appropriately fine-tuned, can achieve comparable results, thereby presenting a viable alternative for extreme speech classification tasks.

Abstract

The world as we know it is constantly shaped by technological innovation, much of which is enabled by advances in semiconductor devices. These devices now play a pivotal role in nearly every field from consumer electronics to automotive systems and critical infrastructure. As their deployment becomes more widespread and their roles increasingly critical, the need for accurate, reliable, and energy efficient operation without compromising on cost has become a central design challenge. This thesis addresses these concerns by proposing a set of techniques to enhance the robustness, power efficiency, and scalability of current reference circuits. Current References serve as critical building blocks in analog and mixed-signal systems. Hence, their performance directly impacts the efficiency of all the other blocks of the SoC. So designing a robust current reference is critical. Among the many challenges in designing such circuits, achieving ultra-low-power operation while maintaining accuracy, thermal stability, and compact area remains particularly demanding. Conventional approaches often suffer from significant area overhead or exhibit substantial performance drift across temperature and process variations. To address these limitations this work presents a 250pA gate-leakage-based current reference that maintains reliable operation across a wide temperature range and process corners. This is achieved through the integration of temperature compensation and autocalibration techniques. The proposed architecture delivers ultra low power current reference over wide temperature range across process corners while occupying minimal silicon area, making it well suited for advanced sensors and IoT appliations. Building on the idea of calibration, this thesis then proposes an autocalibration algorithm applicable to a poly resistance based current references as an alternative to traditional trimming methods thereby achieving cost effective solution.It is seen that the variations are reduced to ±3.5% post calibration while compared to ±30% pre calibration. Then the focus is on machine learning (ML)-based optimization techniques for analog circuit design. The methodology uses data-driven surrogate modeling to explore the high-dimensional design space efficiently. The approach is verified on a double balanced gilbert mixer cell to calibrate for P1dB and IIP3 parameters. Together this works supports the advancement of analog design for efficient performance.

Abstract

The rapid advancement of Transformer-based models has transformed natural language processing (NLP), yet their application to low-resource Indic languages remains underexplored, limiting equitable access to digital knowledge. This thesis investigates the encoding capabilities and robustness of multilingual Transformer models for Indic languages, culminating in practical NLP applications to enhance knowledge accessibility. By addressing linguistic diversity and digital inclusion, the study aligns with the urgent need to bridge information gaps for millions of Indic language speakers, particularly in India, where regional languages like Telugu face significant resource constraints. Our work begins by introducing INDICSENTEVAL, a novel dataset with approximately 47,000 sentences across six Indic languages—Hindi, Telugu, Marathi, Kannada, Urdu, and Malayalam. This dataset enables a comprehensive probing analysis of nine multilingual Transformer models (seven universal, two Indic-specific) across eight linguistic properties, including surface, syntactic, and semantic features. The findings reveal that while models encode English properties consistently, performance for Indic languages varies, with Indic-specific models outperforming universal ones. Extending this analysis, our study evaluates model robustness against 13 perturbations, such as word dropping and shuffling, finding that universal models exhibit greater resilience, particularly under noun and verb perturbations. These insights, detailed in Chapters 3 and 4, highlight model-specific strengths and weaknesses, informing robust NLP system design. Moreover, the development of IndicSentEval and the resulting insights into model behavior underscore the critical role of digitized linguistic data in enabling effective downstream NLP tasks for low-resource languages. Building on these findings, the thesis develops a practical application to enhance Telugu Wikipedia, addressing the knowledge access gap for 95.7 million Telugu speakers. Leveraging NLP techniques like translation, transliteration, and template generation, the study generates 8,929 high-quality moviedomain articles, adhering to Wikipedia’s standards for word count, images, and infoboxes. This contribution, detailed in Chapter 5, demonstrates the real-world impact of robust NLP systems, fostering digital inclusion for low-resource language communitie

Abstract

The proliferation of user-generated content on microblogging platforms like Twitter and via SMS has created a wealth of short-form, opinionated data. However, the noisy, concise, and context-dependent nature of this text presents significant challenges for traditional sentiment analysis techniques. This thesis addresses the task of sentiment polarity classification for short expressions by conducting a systematic, two-stage investigation into feature engineering, moving from surface-level patterns to deep syntactic analysis. The initial phase of our research investigates the effectiveness of three feature categories: (1) phraselevel prior polarity features derived from established resources like SentiWordNet and custom-built lexicons for emoticons and interjections; (2) Part-of-Speech (POS) tag pattern features; and (3) noisy data-specific features. Using a Naive Bayes classifier on the SemEval-2013 dataset, our initial experiments revealed that while these features provide a strong baseline, they have limitations. Specifically, while POS tag patterns offered a boost for subjectivity classification (distinguishing objective vs. subjective text), their impact on polarity classification (positive, negative, neutral) was minimal. This highlighted a key weakness: an inability to robustly model complex contextual phenomena like negation and modification, resulting in lower performance for negative expressions. To address these shortcomings, the second and primary phase of this thesis introduces a novel approach centered on Phrase Dependency Parsing. By extending traditional word-level dependency parsing to the phrase level, we combine the relational benefits of dependency graphs with the structural awareness of constituency trees. This method allows us to first define a precise contextual boundary for a given expression within its parent sentence. From this pruned, context-aware sub-tree, we extract a new set of powerful Syntactic Dependency Features, including a Phrase Path, a Sentiment Phrase Path, and a crucial Subject Negation feature. By integrating these syntactic features and employing a more powerful Support Vector Machine (SVM) classifier, our system’s performance improved dramatically. The average F1-score on the Twitter dataset jumped from 0.744 to 0.863. Most notably, the F-measure for the challenging negative class saw a substantial increase from 0.676 to 0.834. This result strongly validates our central hypothesis: a deep understanding of syntactic structure is essential for accurately classifying sentiment in short, noisy text. This work demonstrates that moving beyond surface-level features to incorporate linguisticallyinformed, dependency-based features is a critical step in overcoming the challenges of contextual polarity

Abstract

This thesis investigates two aspects of how humans engage with visual content: how it is described and how it is remembered. The first part focusses on Audio Descriptions (ADs), which convey essential on-screen information, allowing visually impaired audiences to follow videos. To be effective, ADs must form a coherent sequence that helps listeners visualise the unfolding scene, rather than describing isolated moments. However, most automatic methods generate each AD independently, often resulting in repetitive, incoherent descriptions. To address this, we propose a training-free method, CoherentAD, that first generates multiple candidate descriptions for each AD time interval, and then performs auto-regressive selection across the sequence to form a coherent and informative narrative. To evaluate AD sequences holistically, we introduce a sequence-level metric, StoryRecall, which measures how well the predicted ADs convey the ground truth narrative, alongside repetition metrics that capture the redundancy across consecutive AD outputs. Our method produces coherent AD sequences with enhanced narrative understanding, outperforming prior approaches that rely on independent generations. The second part investigates Video Memorability. Understanding what makes a video memorable has important applications in advertising and education technology. Towards this goal, we investigate spatio-temporal attention mechanisms underlying video memorability. Different from previous works that fuse multiple features, we adopt a simple CNN+Transformer architecture that enables analysis of spatio-temporal attention while matching state-of-the-art (SoTA) performance on video memorability prediction. We compare model attention against human gaze fixations collected through a small-scale eye-tracking study where humans perform the video memory task. We uncover the following insights: (i) Quantitative saliency metrics show that our model, trained only to predict a memorability score, exhibits similar spatial attention patterns to human gaze, especially for more memorable videos. (ii) The model assigns greater importance to initial frames in a video, mimicking human attention patterns. (iii) Panoptic segmentation reveals that both (model and humans) assign a greater share of attention to things and less attention to stuff as compared to their occurrence probability

Abstract

This thesis investigates how entities are shaped by the discourse in which they appear. Rather than being isolated mentions, entities take on meaning through their interactions with the surrounding context, both in terms of reference and interpretation: the identity of an entity can shift depending on how it is invoked, and the narrative framing around it can significantly influence how it is perceived by a reader. At its core, the research addresses these two complementary dimensions of entity analysis: the challenge of tracking entity references across text (coreference resolution) and the interpretive challenge of understanding how entities are characterized through their framing within the text (entity role classification). While these tasks represent distinct technical problems, they are unified by their contribution to richer discourse comprehension and their shared goal of moving NLP systems beyond surface-level language processing toward deeper semantic understanding. The first part of this research focuses on coreference resolution for South Asian languages, addressing a significant gap in current NLP research where these morphologically rich and syntactically diverse languages remain underrepresented. To overcome the lack of suitable evaluation resources, we develop mGAP, a novel dataset containing coreference annotations across 27 South Asian languages, created through a dedicated translation and alignment pipeline from English reference data. This dataset provided the foundation for investigating cross-lingual transfer learning strategies designed to address the low-resource nature of many of these languages. Through systematic experimentation, we examine which languages serve as effective source languages for training when target language data is scarce, offering insights into making use of linguistic similarities for enhanced cross-lingual adaptation. The second part shifts focus to entity framing in news media, where the research question evolves from identifying what an entity refers to, to understanding what narrative role that entity plays within the text. The framing of the text is pivotal in how a reader would perceive the entity being described. For this task, we developed a two-stage pipeline for classifying entities into roles such as protagonist, antagonist, and innocent, as well as more fine-grained narrative categories. The pipeline first employs entity-centric summarization to extract relevant contextual information surrounding target entities, followed by classification using a DeBERTa v3-based model enhanced with contrastive learning techniques. This approach addresses the significant challenge of determining how entities are portrayed based on surrounding textual cues, contributing further to computational methods for analyzing media bias and narrative construction.

Abstract

India is experiencing an innovation upswing in many fields such as: pharmaceuticals, agriculture, space technology and especially in the area of IT/computer science. The Indian IT industry has a lot to offer: lower human resource costs, large talent pool, favourable government policies, access to capital, and advanced infrastructure that makes India the cost-effective Silicon Valley of the world. It also hosts a significant number of world-class research institutions producing handsome volumes of deep-tech computer science research. However, it has been observed that much of this research is unable to productise and become usable for markets in the real world This has resulted in a high volume of research with potential to disrupt markets but fail to carry through the ‘lab to land’ journey for a research output. This thesis is a study to understand the translation process of a research as a marketable product and to identify factors that impede or enable this translation. The productisation window for any relevant computer science research work is largely time sensitive. [1] Recent explosion in AI adoption fueled with a digital infrastructure has made many deep-tech Computer Science research areas to get production ready very fast. It has an associated shelf life during which it should be explored and leveraged for any sort of translational or productisation activity. Therefore, as a community engaging closely with research problems bridging disconnections between research labs and industry/market requires critical and timely interventions. While there are many research translation models available as a concept or practice, many are ineffective in explaining a seamless / smooth transition of any research work into an industry-consumable technology or product. Based on tacit knowledge & existing models of translation it is apparent to observe that due to the nature of the research ecosystem, its processes and the associated stakeholders within it, identifying potential translation flow is far from trivial. In order to understand the entire journey of a potential research translation, this thesis proposes a productisation flow that considers all potential channels through which a research work in computer science can transition into a market relevant product or an industry consumable technology. This thesis will cover the productisation flow in detail in the later chapters. In this study, we focus on understanding this gap in productisation from the perspective of research scholars who actually work on these projects. After an extensive literature study, the thesis focuses on highlighting the gap in considering ‘researchers’ as an active stakeholder in this entire productisation process. As a result, this thesis proposes a productisation flow of deep-tech research from lab to market translations that highlights the participation of researchers across various translational channels and their stages. This qualifies the need to understand reasons why research flow from lab to land is enabled or inhibited wrt researchers.