Vadivel Abishethvarman

Mr. Vadivel Abishethvarman

orcid
Never stop learning

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 May 2026 in Machine Learning
Verified
Journal Article
Open access
PDF file
PJPratik Jalan
Pratik Jalan
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
BCBhavik Chandna
Bhavik Chandna
UNUsman Naseem
Usman Naseem
Large Language Models (LLM) have demonstrated remarkable capabilities across various applications, but their deployment raises critical safety concerns as potential misuse poses significant societal risks. This survey reviews the end-to-end security and safety pipeline of LLMs, focusing on the interaction between users and model responses. We categorize the system into five key components: attacks, defenses, safety alignment, metrics and guarding mechanisms. Attacks involve crafting adversarial inputs to exploit model vulnerabilities. Defenses act as countermeasures, aiming to detect and prevent such inputs before processing. Safety alignment ensures that, even when attacks reach the model, its responses remain consistent with ethical and policy-aligned behavior. Guarding mechanisms operate post-response to flag, filter, or block unsafe outputs. Each of these stages is subject to rigorous evaluation metrics to assess their effectiveness, robustness, and limitations. As LLMs evolve toward more general-purpose intelligence, these safety considerations become increasingly critical for the development of robust and trustworthy AI systems. Finally, we highlight open challenges and future research directions to advance the security and alignment of LLMs.

4 December 2025 in Information
Verified
Journal Article
Open access
PDF file
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
FSFariza Sabrina
Fariza Sabrina
WKWing Hing Paul Kwan
Wing Hing Paul Kwan
Large Language Models (LLMs) are emerging technologies and a growing research trend in Artificial General Intelligence (AGI), which envisions a future where machines can think and learn like humans across a wide range of tasks. Information generated by LLMs is essentially the prediction of next tokens in Natural Language Processing (NLP) tasks. However, the generated content is always subject to issues of truthfulness and hallucinations. The information and knowledge integrity of LLM-generated content therefore remains subjective. Exploring recent literature on the integrity of LLMs in a systematic manner is both timely and essential. Moreover, ensuring the reliability of LLMs in real-world applications is critical. Various approaches have been explored to promote information and knowledge integrity in LLMs, including adversarial training, data augmentation, and calibration methods. However, beyond these techniques, other strategies also contribute to maintaining knowledge integrity. This paper specifically focuses on three such approaches: knowledge distillation, semantic integrity, and provenance tracking, which play essential roles in ensuring that LLMs generate accurate, consistent, and trustworthy information. Knowledge distillation enhances model efficiency by transferring knowledge from larger models to smaller ones while preserving essential learning without compromising knowledge integrity. This reduces hallucinations. Semantic integrity safeguards consistency and strengthens the robustness of generated outputs. It is concurrently checking the meaningfulness of the outputs with the context. Provenance tracking improves transparency and trustworthiness through mechanisms such as data lineage and explainability, thereby ensuring the credibility of the LLM-generated responses. This review suggests that knowledge distillation, semantic integrity, and provenance tracking can enhance the reliability of LLM outputs, with prior studies reporting reductions in hallucination rates, improvements in robustness, and gains in factual consistency.

8 October 2024 in Sri Lanka Journal of Social Sciences and Humanities
Verified
Journal Article
Open access
PDF file
Banujan KuhaneswaranBanujan Kuhaneswaran
Banujan Kuhaneswaran
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
AWAshansa Wijeratne
Ashansa Wijeratne
NRNirubikaa Ravikumar
Nirubikaa Ravikumar
SKSamantha Kumara
Samantha Kumara
TVThanosan Vijayanandan
Thanosan Vijayanandan
In the rapidly evolving landscape of Artificial Intelligence (AI), platforms like ChatGPT are reshaping the educational domain, prompting deeper explorations into the nature and depth of this intersection. This study aimed to systematically uncover the prevailing sentiments, concerns and discussions on Twitter surrounding ChatGPT’s role in education. Through an extensive data collection process, over 3.8 million tweets were initially gathered, followed by rigorous refining processes that included expert-driven tweet labelling and subsequent classification using Machine Learning (ML) and deep learning models. The cleaned dataset underwent a series of preprocessing steps and feature extraction and was ultimately subjected to Latent Dirichlet Allocation (LDA) for topic modelling. Our findings unveiled 15 distinct topics that broadly spanned common discussions, AI implementation, and its potential impacts. The data’s visualisation using t-distributed stochastic neighbour embedding (t-SNE) showcased a dense central clustering of these topics. In conclusion, our research underscores the multi-faceted dialogues on AI, particularly ChatGPT, in education, emphasising the pressing need for continued discourse and research as AI tools further integrate into our educational paradigms.

21 February 2024
Verified
Conference Paper
Closed access
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
KBKuhaneswaran Banujan
Kuhaneswaran Banujan
BKB.T.G.S Kumara
B.T.G.S Kumara
NRNirubikaa Ravikumar
Nirubikaa Ravikumar
Discerning truth from deception requires innovative tools in an era besieged by misinformation. Fake news is a form of misleading information delivered as reliable news. It is mainly made up of intentionally manipulating public opinion, decision bias, and exploiting general credulity. Various supervised learning-based fake news detection and classifications only focus on the semantic values due to their subjectivity in the current context. Sentiment analysis is also a common interpretation mechanism of information mining. This research explores the role of sentiment analysis in the context of fake news detection. Sentiment analysis, a facet of natural language processing, delves into the nuanced sentiments within the text, unveiling hidden narratives and differentiating between authentic reporting and carefully crafted deception. The study employs various sentiment reasoning models, including VADER, TextBlob, and Roberta, focusing on comparative analysis. The VADER model, complemented by a Bag of Words approach, emerges as the most effective, outperforming, compared to TextBlob and RoBERTa. Especially BERT sentiment reasoner model underperformed with uneven distribution in this fake news domain. The best VADER sentiments are taken for the transitions of sequential neural network architectures such as Simple Neural Network (SNN), Convolutional Neural Network (CNN), and Recurrent Neural Network (RNN) featuring Long Short-Term Memory (LSTM). The LSTM model captures intricate patterns, dependencies, and contextual information. Evaluation metrics, encompassing accuracy, precision, recall, and F1 score, provide a comprehensive understanding of models' performance. The RNN LSTM performed well compared to the other sequential models. It primely emphasised the multifaceted landscape of sentiment analysis, presenting a robust strategy for combating the proliferation of fake news in an information-abundant age. Combining Sentiment Reasoning in VADER and RNN LSTM presents a robust strategy for fake news detection as a novel approach in the evolving field of misinformation detection. The study discusses the importance of sentiment reasoning for the fake news datasets contributing in terms of the advancement in fake news classification and detection.

25 October 2023
Verified
Conference Paper
Closed access
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
KBKuhaneswaran Banujan
Kuhaneswaran Banujan
AWAshansa Wijeratne
Ashansa Wijeratne
RNRavikumar Nirubikaa
Ravikumar Nirubikaa
SKSamantha Kumara
Samantha Kumara
SPSenthan Prasanth
Senthan Prasanth
The study comprehensively analysed public sentiment towards ChatGPT, a language processing AI tool, using approximately 135,000 tweets from an initial dataset of 236,275 entries. After cleaning and pre-processing the data using basic NLP techniques, Machine Learning (ML) models—Long Short-Term Memory (LSTM), Decision Tree (DT), and Support Vector Machine (SVM)—were applied to classify the tweets into positive, negative, and neutral categories. The LSTM model demonstrated superior performance across all evaluation metrics—accuracy, precision, recall, and F1 score—for all tweet categories. Word clouds generated for each category revealed that the most frequent words in positive tweets included ‘student,’ 'education,’ 'research,’ and ‘teacher,’ indicating a strong association of ChatGPT with educational and research contexts. Similarly, ‘AI,’ ‘student,’ 'research,’ ‘course,’ and ‘education’ dominated neutral and negative tweets, highlighting prevalent discussion topics. However, the study had limitations, including dependency on a single dataset, potential loss of contextual information during pre-processing, and a focus on English tweets only. Despite these limitations, the study provides valuable insights into the public’s perception of ChatGPT, underscoring its significant association with education and research and the effectiveness of the LSTM model in classifying tweets. The findings suggest that future research should incorporate diverse data sources and languages better to understand global sentiment towards ChatGPT and similar AI tools.

11 November 2025
Verified
Book Chapter
Closed access
PDF file
GBGobihanath Balasubramaniam
Gobihanath Balasubramaniam
Vadivel AbishethvarmanVadivel Abishethvarman
Vadivel Abishethvarman
B. T. G. Samantha KumaraB. T. G. Samantha Kumara
B. T. G. Samantha Kumara
SPSenthan Prasanth
Senthan Prasanth
Banujan KuhaneswaranBanujan Kuhaneswaran
Banujan Kuhaneswaran