12 papers
Where To Look? : Causal Tracing of Vision Encoders in VLM
Naren Kumar S, Tirth Bhatt, Mayank Singh
Vision-language models can describe an image with remarkable accuracy, yet a more fundamental question remains unanswered: what visual information actually drives their answers? In…
Task-Conditional Flow Matching for Balanced Multilingual Text Embedding Adaptation
Tirth Bhatt, Naren Kumar S, Mayank Singh
Multilingual text embedding models are commonly adapted using a single training objective across diverse tasks, despite different tasks requiring fundamentally different optimizati…
Gurukul AI: An Interactive AI-Driven Educational Platform for Indian Education System
Isha Narang, Sneh Gosai, Mayank Singh
Recent advances in large language models (LLMs) like ChatGPT and LLaMA have transformed AI-driven education, but these systems are predominantly trained on Western-centric data, ma…
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models
Arya Shah, Himanshu Beniwal, Mayank Singh +1
Safety-aligned large language models often exhibit sycophancy, which is the tendency to affirm users' opinions regardless of factual accuracy. Although well-studied in English, its…
Where Does Toxicity Live? Mechanistic Localization and Targeted Suppression in Language Models
Himanshu Beniwal, Mayank Singh
Large language models frequently generate toxic, hateful, or harmful content, yet existing mitigation methods rely on costly retraining or output-level filtering with no mechanisti…
One Instruction Does Not Fit All: How Well Do Embeddings Align Personas and Instructions in Low-Resource Indian Languages?
Arya Shah, Himanshu beniwal, Mayank Singh
Aligning multilingual assistants with culturally grounded user preferences is essential for serving India's linguistically diverse population of over one billion speakers across mu…
UNITYAI-GUARD: Pioneering Toxicity Detection Across Low-Resource Indian Languages
Himanshu Beniwal, Reddybathuni Venkat, Rohit Kumar +7
This work introduces UnityAI-Guard, a framework for binary toxicity classification targeting low-resource Indian languages. While existing systems predominantly cater to high-resou…
COMI-LINGUA: Expert Annotated Large-Scale Dataset for Multitask NLP in Hindi-English Code-Mixing
Rajvee Sheth, Himanshu Beniwal, Mayank Singh
We introduce COMI-LINGUA, the largest manually annotated Hindi-English code-mixed dataset, comprising 125K+ high-quality instances across five core NLP tasks: Matrix Language Ident…
Char-mander Use mBackdoor! A Study of Cross-lingual Backdoor Attacks in Multilingual LLMs
Himanshu Beniwal, Sailesh Panda, Birudugadda Srivibhav +1
We explore \textbf{C}ross-lingual \textbf{B}ackdoor \textbf{AT}tacks (X-BAT) in multilingual Large Language Models (mLLMs), revealing how backdoors inserted in one language can aut…
COMMENTATOR: A Code-mixed Multilingual Text Annotation Framework
Rajvee Sheth, Shubh Nisar, Heenaben Prajapati +2
As the NLP community increasingly addresses challenges associated with multilingualism, robust annotation tools are essential to handle multilingual datasets efficiently. In this p…
Cross-lingual Editing in Multilingual Language Models
Himanshu Beniwal, Kowsik Nandagopan D, Mayank Singh
The training of large language models (LLMs) necessitates substantial data and computational resources, and updating outdated LLMs entails significant efforts and resources. While…
Remember This Event That Year? Assessing Temporal Information and Reasoning in Large Language Models
Himanshu Beniwal, Dishant Patel, Kowsik Nandagopan D +3
Large Language Models (LLMs) are increasingly ubiquitous, yet their ability to retain and reason about temporal information remains limited, hindering their application in real-wor…