arXiv subjects

Vibhor Agarwal

Publications and source records attributed to Vibhor Agarwal.

15 recordsLinked to original sources

arXiv2026

Frictionless Love: Associations Between AI Companion Roles and Behavioral Addiction

AI companion chatbots increasingly shape how people seek social and emotional connection, sometimes substituting for relationships with romantic partners, friends, teachers, or even therapists. When these systems adopt those metaphorical roles, they are not neutral: such roles structure people's ways of interacting, distribute perceived AI harms and benefits, and may reflect behavioral addiction signs. Yet these role-dependent risks remain poorly understood. We analyze 248,830 posts from seven prominent Reddit communities describing interactions with AI companions. We identify ten recurring metaphorical roles (for example, soulmate, philosopher, and coach) and show that each role supports distinct ways of interacting. We then extract the perceived AI harms and AI benefits associated with these role-specific interactions and link them to behavioral addiction signs, all of which has been inferred from the text in the posts. AI soulmate companions are associated with romance-centered ways of interacting, offering emotional support but also introducing emotional manipulation and distress, culminating in strong attachment. In contrast, AI coach and guardian companions are associated with practical benefits such as personal growth and task support, yet are nonetheless more frequently associated with behavioral addiction signs such as daily life disruptions and damage to offline relationships. These findings show that metaphorical roles are a central ethical design concern for responsible AI companions.

cs.CY↗

arXiv2025

MedHalu: Hallucinations in Responses to Healthcare Queries by Large Language Models

Large language models (LLMs) are starting to complement traditional information seeking mechanisms such as web search. LLM-powered chatbots like ChatGPT are gaining prominence among the general public. AI chatbots are also increasingly producing content on social media platforms. However, LLMs are also prone to hallucinations, generating plausible yet factually incorrect or fabricated information. This becomes a critical problem when laypeople start seeking information about sensitive issues such as healthcare. Existing works in LLM hallucinations in the medical domain mainly focus on testing the medical knowledge of LLMs through standardized medical exam questions which are often well-defined and clear-cut with definitive answers. However, these approaches may not fully capture how these LLMs perform during real-world interactions with patients. This work conducts a pioneering study on hallucinations in LLM-generated responses to real-world healthcare queries from patients.We introduce MedHalu, a novel medical hallucination benchmark featuring diverse health-related topics and hallucinated responses from LLMs, with detailed annotation of the hallucination types and text spans. We also propose MedHaluDetect, a comprehensive framework for evaluating LLMs' abilities to detect hallucinations. Furthermore, we study the vulnerability to medical hallucinations among three groups -- medical experts, LLMs, and laypeople. Notably, LLMs significantly underperform human experts and, in some cases, even laypeople in detecting medical hallucinations. To improve hallucination detection, we propose an expert-in-the-loop approach that integrates expert reasoning into LLM inputs, significantly improving hallucination detection for all LLMs, including a 6.3% macro-F1 improvement for GPT-4. Our code and dataset are available at https://netsys.surrey.ac.uk/datasets/medhalu/.

cs.CL↗

arXiv2025

CodeMirage: Hallucinations in Code Generated by Large Language Models

Large Language Models (LLMs) have shown promising potentials in program generation and no-code automation. However, LLMs are prone to generate hallucinations, i.e., they generate text which sounds plausible but is incorrect. Although there has been a recent surge in research on LLM hallucinations for text generation, similar hallucination phenomenon can happen in code generation. Sometimes the generated code can have syntactical or logical errors as well as more advanced issues like security vulnerabilities, memory leaks, etc. Given the wide adaptation of LLMs to enhance efficiency in code generation and development in general, it becomes imperative to investigate hallucinations in code generation. To the best of our knowledge, this is the first attempt at studying hallucinations in the code generated by LLMs. We start by introducing the code hallucination definition and a comprehensive taxonomy of code hallucination types. We propose the first benchmark CodeMirage dataset for code hallucinations. The benchmark contains 1,137 GPT-3.5 generated hallucinated code snippets for Python programming problems from two base datasets - HumanEval and MBPP. We then propose the methodology for code hallucination detection and experiment with open source LLMs such as CodeLLaMA as well as OpenAI's GPT-3.5 and GPT-4 models using one-shot prompt. We find that GPT-4 performs the best on HumanEval dataset and gives comparable results to the fine-tuned CodeBERT baseline on MBPP dataset. Towards the end, we discuss various mitigation strategies for code hallucinations and conclude our work.

cs.SE↗

arXiv2025

Conversation Kernels: A Flexible Mechanism to Learn Relevant Context for Online Conversation Understanding

Understanding online conversations has attracted research attention with the growth of social networks and online discussion forums. Content analysis of posts and replies in online conversations is difficult because each individual utterance is usually short and may implicitly refer to other posts within the same conversation. Thus, understanding individual posts requires capturing the conversational context and dependencies between different parts of a conversation tree and then encoding the context dependencies between posts and comments/replies into the language model. To this end, we propose a general-purpose mechanism to discover appropriate conversational context for various aspects about an online post in a conversation, such as whether it is informative, insightful, interesting or funny. Specifically, we design two families of Conversation Kernels, which explore different parts of the neighborhood of a post in the tree representing the conversation and through this, build relevant conversational context that is appropriate for each task being considered. We apply our developed method to conversations crawled from slashdot.org, which allows users to apply highly different labels to posts, such as 'insightful', 'funny', etc., and therefore provides an ideal experimental platform to study whether a framework such as Conversation Kernels is general-purpose and flexible enough to be adapted to disparately different conversation understanding tasks.

cs.CL↗

arXiv2024

Decentralised Moderation for Interoperable Social Networks: A Conversation-based Approach for Pleroma and the Fediverse

The recent development of decentralised and interoperable social networks (such as the "fediverse") creates new challenges for content moderators. This is because millions of posts generated on one server can easily "spread" to another, even if the recipient server has very different moderation policies. An obvious solution would be to leverage moderation tools to automatically tag (and filter) posts that contravene moderation policies, e.g. related to toxic speech. Recent work has exploited the conversational context of a post to improve this automatic tagging, e.g. using the replies to a post to help classify if it contains toxic speech. This has shown particular potential in environments with large training sets that contain complete conversations. This, however, creates challenges in a decentralised context, as a single conversation may be fragmented across multiple servers. Thus, each server only has a partial view of an entire conversation because conversations are often federated across servers in a non-synchronized fashion. To address this, we propose a decentralised conversation-aware content moderation approach suitable for the fediverse. Our approach employs a graph deep learning model (GraphNLI) trained locally on each server. The model exploits local data to train a model that combines post and conversational information captured through random walks to detect toxicity. We evaluate our approach with data from Pleroma, a major decentralised and interoperable micro-blogging network containing 2 million conversations. Our model effectively detects toxicity on larger instances, exclusively trained using their local post information (0.8837 macro-F1). Our approach has considerable scope to improve moderation in decentralised and interoperable social networks such as Pleroma or Mastodon.

cs.CY↗

arXiv2024

"Which LLM should I use?": Evaluating LLMs for tasks performed by Undergraduate Computer Science Students

This study evaluates the effectiveness of various large language models (LLMs) in performing tasks common among undergraduate computer science students. Although a number of research studies in the computing education community have explored the possibility of using LLMs for a variety of tasks, there is a lack of comprehensive research comparing different LLMs and evaluating which LLMs are most effective for different tasks. Our research systematically assesses some of the publicly available LLMs such as Google Bard, ChatGPT(3.5), GitHub Copilot Chat, and Microsoft Copilot across diverse tasks commonly encountered by undergraduate computer science students in India. These tasks include code explanation and documentation, solving class assignments, technical interview preparation, learning new concepts and frameworks, and email writing. Evaluation for these tasks was carried out by pre-final year and final year undergraduate computer science students and provides insights into the models' strengths and limitations. This study aims to guide students as well as instructors in selecting suitable LLMs for any specific task and offers valuable insights on how LLMs can be used constructively by students and instructors.

cs.CY↗

arXiv2023

HateRephrase: Zero- and Few-Shot Reduction of Hate Intensity in Online Posts using Large Language Models

Hate speech has become pervasive in today's digital age. Although there has been considerable research to detect hate speech or generate counter speech to combat hateful views, these approaches still cannot completely eliminate the potential harmful societal consequences of hate speech -- hate speech, even when detected, can often not be taken down or is often not taken down enough; and hate speech unfortunately spreads quickly, often much faster than any generated counter speech. This paper investigates a relatively new yet simple and effective approach of suggesting a rephrasing of potential hate speech content even before the post is made. We show that Large Language Models (LLMs) perform well on this task, outperforming state-of-the-art baselines such as BART-Detox. We develop 4 different prompts based on task description, hate definition, few-shot demonstrations and chain-of-thoughts for comprehensive experiments and conduct experiments on open-source LLMs such as LLaMA-1, LLaMA-2 chat, Vicuna as well as OpenAI's GPT-3.5. We propose various evaluation metrics to measure the efficacy of the generated text and ensure the generated text has reduced hate intensity without drastically changing the semantic meaning of the original text. We find that LLMs with a few-shot demonstrations prompt work the best in generating acceptable hate-rephrased text with semantic meaning similar to the original text. Overall, we find that GPT-3.5 outperforms the baseline and open-source models for all the different kinds of prompts. We also perform human evaluations and interestingly, find that the rephrasings generated by GPT-3.5 outperform even the human-generated ground-truth rephrasings in the dataset. We also conduct detailed ablation studies to investigate why LLMs work satisfactorily on this task and conduct a failure analysis to understand the gaps.

cs.CL↗

arXiv2023

GASCOM: Graph-based Attentive Semantic Context Modeling for Online Conversation Understanding

Online conversation understanding is an important yet challenging NLP problem which has many useful applications (e.g., hate speech detection). However, online conversations typically unfold over a series of posts and replies to those posts, forming a tree structure within which individual posts may refer to semantic context from higher up the tree. Such semantic cross-referencing makes it difficult to understand a single post by itself; yet considering the entire conversation tree is not only difficult to scale but can also be misleading as a single conversation may have several distinct threads or points, not all of which are relevant to the post being considered. In this paper, we propose a Graph-based Attentive Semantic COntext Modeling (GASCOM) framework for online conversation understanding. Specifically, we design two novel algorithms that utilise both the graph structure of the online conversation as well as the semantic information from individual posts for retrieving relevant context nodes from the whole conversation. We further design a token-level multi-head graph attention mechanism to pay different attentions to different tokens from different selected context utterances for fine-grained conversation context modeling. Using this semantic conversational context, we re-examine two well-studied problems: polarity prediction and hate speech detection. Our proposed framework significantly outperforms state-of-the-art methods on both tasks, improving macro-F1 scores by 4.5% for polarity prediction and by 5% for hate speech detection. The GASCOM context weights also enhance interpretability.

cs.CL↗

arXiv2023

AI in the Gray: Exploring Moderation Policies in Dialogic Large Language Models vs. Human Answers in Controversial Topics

The introduction of ChatGPT and the subsequent improvement of Large Language Models (LLMs) have prompted more and more individuals to turn to the use of ChatBots, both for information and assistance with decision-making. However, the information the user is after is often not formulated by these ChatBots objectively enough to be provided with a definite, globally accepted answer. Controversial topics, such as "religion", "gender identity", "freedom of speech", and "equality", among others, can be a source of conflict as partisan or biased answers can reinforce preconceived notions or promote disinformation. By exposing ChatGPT to such debatable questions, we aim to understand its level of awareness and if existing models are subject to socio-political and/or economic biases. We also aim to explore how AI-generated answers compare to human ones. For exploring this, we use a dataset of a social media platform created for the purpose of debating human-generated claims on polemic subjects among users, dubbed Kialo. Our results show that while previous versions of ChatGPT have had important issues with controversial topics, more recent versions of ChatGPT (gpt-3.5-turbo) are no longer manifesting significant explicit biases in several knowledge areas. In particular, it is well-moderated regarding economic aspects. However, it still maintains degrees of implicit libertarian leaning toward right-winged ideals which suggest the need for increased moderation from the socio-political point of view. In terms of domain knowledge on controversial topics, with the exception of the "Philosophical" category, ChatGPT is performing well in keeping up with the collective human level of knowledge. Finally, we see that sources of Bing AI have slightly more tendency to the center when compared to human answers. All the analyses we make are generalizable to other types of biases and domains.

cs.LG↗

arXiv2023

AnnoBERT: Effectively Representing Multiple Annotators' Label Choices to Improve Hate Speech Detection

Supervised approaches generally rely on majority-based labels. However, it is hard to achieve high agreement among annotators in subjective tasks such as hate speech detection. Existing neural network models principally regard labels as categorical variables, while ignoring the semantic information in diverse label texts. In this paper, we propose AnnoBERT, a first-of-its-kind architecture integrating annotator characteristics and label text with a transformer-based model to detect hate speech, with unique representations based on each annotator's characteristics via Collaborative Topic Regression (CTR) and integrate label text to enrich textual representations. During training, the model associates annotators with their label choices given a piece of text; during evaluation, when label information is not available, the model predicts the aggregated label given by the participating annotators by utilising the learnt association. The proposed approach displayed an advantage in detecting hate speech, especially in the minority class and edge cases with annotator disagreement. Improvement in the overall performance is the largest when the dataset is more label-imbalanced, suggesting its practical value in identifying real-world hate speech, as the volume of hate speech in-the-wild is extremely small on social media, when compared with normal (non-hate) speech. Through ablation studies, we show the relative contributions of annotator embeddings and label text to the model performance, and tested a range of alternative annotator embeddings and label text combinations.

cs.CL↗

arXiv2022

A Graph-Based Context-Aware Model to Understand Online Conversations

Online forums that allow for participatory engagement between users have been transformative for the public discussion of many important issues. However, such conversations can sometimes escalate into full-blown exchanges of hate and misinformation. Existing approaches in natural language processing (NLP), such as deep learning models for classification tasks, use as inputs only a single comment or a pair of comments depending upon whether the task concerns the inference of properties of the individual comments or the replies between pairs of comments, respectively. But in online conversations, comments and replies may be based on external context beyond the immediately relevant information that is input to the model. Therefore, being aware of the conversations' surrounding contexts should improve the model's performance for the inference task at hand. We propose GraphNLI, a novel graph-based deep learning architecture that uses graph walks to incorporate the wider context of a conversation in a principled manner. Specifically, a graph walk starts from a given comment and samples "nearby" comments in the same or parallel conversation threads, which results in additional embeddings that are aggregated together with the initial comment's embedding. We then use these enriched embeddings for downstream NLP prediction tasks that are important for online conversations. We evaluate GraphNLI on two such tasks - polarity prediction and misogynistic hate speech detection - and found that our model consistently outperforms all relevant baselines for both tasks. Specifically, GraphNLI with a biased root-seeking random walk performs with a macro-F1 score of 3 and 6 percentage points better than the best-performing BERT-based baselines for the polarity prediction and hate speech detection tasks, respectively.

cs.CL↗

arXiv2022

GraphNLI: A Graph-based Natural Language Inference Model for Polarity Prediction in Online Debates

Online forums that allow participatory engagement between users have been transformative for public discussion of important issues. However, debates on such forums can sometimes escalate into full blown exchanges of hate or misinformation. An important tool in understanding and tackling such problems is to be able to infer the argumentative relation of whether a reply is supporting or attacking the post it is replying to. This so called polarity prediction task is difficult because replies may be based on external context beyond a post and the reply whose polarity is being predicted. We propose GraphNLI, a novel graph-based deep learning architecture that uses graph walk techniques to capture the wider context of a discussion thread in a principled fashion. Specifically, we propose methods to perform root-seeking graph walks that start from a post and captures its surrounding context to generate additional embeddings for the post. We then use these embeddings to predict the polarity relation between a reply and the post it is replying to. We evaluate the performance of our models on a curated debate dataset from Kialo, an online debating platform. Our model outperforms relevant baselines, including S-BERT, with an overall accuracy of 83%.

cs.CL↗

arXiv2022

"Way back then": A Data-driven View of 25+ years of Web Evolution

Since the inception of the first web page three decades back, the Web has evolved considerably, from static HTML pages in the beginning to the dynamic web pages of today, from mainly the text-based pages of the 1990s to today's multimedia rich pages, etc. Although much of this is known anecdotally, to our knowledge, there is no quantitative documentation of the extent and timing of these changes. This paper attempts to address this gap in the literature by looking at the top 100 Alexa websites for over 25 years from the Internet Archive or the "Wayback Machine", archive.org. We study the changes in popularity, from Geocities and Yahoo! in the mid-to-late 1990s to the likes of Google, Facebook, and Tiktok of today. We also look at different categories of websites and their popularity over the years and find evidence for the decline in popularity of news and education-related websites, which have been replaced by streaming media and social networking sites. We explore the emergence and relative prevalence of different MIME-types (text vs. image vs. video vs. javascript and json) and study whether the use of text on the Internet is declining.

cs.CY↗

arXiv2021

Under the Spotlight: Web Tracking in Indian Partisan News Websites

India is experiencing intense political partisanship and sectarian divisions. The paper performs, to the best of our knowledge, the first comprehensive analysis on the Indian online news media with respect to tracking and partisanship. We build a dataset of 103 online, mostly mainstream news websites. With the help of two experts, alongside data from the Media Ownership Monitor of the Reporters without Borders, we label these websites according to their partisanship (Left, Right, or Centre). We study and compare user tracking on these sites with different metrics: numbers of cookies, cookie synchronizations, device fingerprinting, and invisible pixel-based tracking. We find that Left and Centre websites serve more cookies than Right-leaning websites. However, through cookie synchronization, more user IDs are synchronized in Left websites than Right or Centre. Canvas fingerprinting is used similarly by Left and Right, and less by Centre. Invisible pixel-based tracking is 50% more intense in Centre-leaning websites than Right, and 25% more than Left. Desktop versions of news websites deliver more cookies than their mobile counterparts. A handful of third-parties are tracking users in most websites in this study. This paper, by demonstrating intense web tracking, has implications for research on overall privacy of users visiting partisan news websites in India.

cs.CY↗

arXiv2021

Differential Tracking Across Topical Webpages of Indian News Media

Online user privacy and tracking have been extensively studied in recent years, especially due to privacy and personal data-related legislations in the EU and the USA, such as the General Data Protection Regulation, ePrivacy Regulation, and California Consumer Privacy Act. Research has revealed novel tracking and personal identifiable information leakage methods that first- and third-parties employ on websites around the world, as well as the intensity of tracking performed on such websites. However, for the sake of scaling to cover a large portion of the Web, most past studies focused on homepages of websites, and did not look deeper into the tracking practices on their topical subpages. The majority of studies focused on the Global North markets such as the EU and the USA. Large markets such as India, which covers 20% of the world population and has no explicit privacy laws, have not been studied in this regard. We aim to address these gaps and focus on the following research questions: Is tracking on topical subpages of Indian news websites different from their homepage? Do third-party trackers prefer to track specific topics? How does this preference compare to the similarity of content shown on these topical subpages? To answer these questions, we propose a novel method for automatic extraction and categorization of Indian news topical subpages based on the details in their URLs. We study the identified topical subpages and compare them with their homepages with respect to the intensity of cookie injection and third-party embeddedness and type. We find differential user tracking among subpages, and between subpages and homepages. We also find a preferential attachment of third-party trackers to specific topics. Also, embedded third-parties tend to track specific subpages simultaneously, revealing possible user profiling in action.

cs.CY↗