Subhrakanta Panda
YOU?
Author Swipe
View article: Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought
Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thought Open
Recent frontier models employ long chain-of-thought reasoning to explore solution spaces in context and achieve stonger performance. While many works study distillation to build smaller yet capable models, most focus on English and little …
View article: FlexDoc: Parameterized Sampling for Diverse Multilingual Synthetic Documents for Training Document Understanding Models
FlexDoc: Parameterized Sampling for Diverse Multilingual Synthetic Documents for Training Document Understanding Models Open
Developing document understanding models at enterprise scale requires large, diverse, and well-annotated datasets spanning a wide range of document types. However, collecting such data is prohibitively expensive due to privacy constraints,…
View article: AccessEval: Benchmarking Disability Bias in Large Language Models
AccessEval: Benchmarking Disability Bias in Large Language Models Open
Large Language Models (LLMs) are increasingly deployed across diverse domains but often exhibit disparities in how they handle real-life queries. To systematically investigate these effects within various disability contexts, we introduce …
View article: Who Gets Left Behind? Auditing Disability Inclusivity in Large Language Models
Who Gets Left Behind? Auditing Disability Inclusivity in Large Language Models Open
Large Language Models (LLMs) are increasingly used for accessibility guidance, yet many disability groups remain underserved by their advice. To address this gap, we present taxonomy aligned benchmark1 of human validated, general purpose a…
View article: Hybrid AI for Responsive Multi-Turn Online Conversations with Novel Dynamic Routing and Feedback Adaptation
Hybrid AI for Responsive Multi-Turn Online Conversations with Novel Dynamic Routing and Feedback Adaptation Open
Retrieval-Augmented Generation (RAG) systems and large language model (LLM)-powered chatbots have significantly advanced conversational AI by combining generative capabilities with external knowledge retrieval. Despite their success, enter…
View article: Adaptive Reuse of Historic Buildings for Preserving the Past
Adaptive Reuse of Historic Buildings for Preserving the Past Open
Making an item from completely new raw materials has a significant negative impact on the environment. This can be significantly reduced by recycling or reusing items for purposes other than those for which they were designed. This logic a…
View article: Secure and efficient ownership verification for deduplicated cloud computing systems
Secure and efficient ownership verification for deduplicated cloud computing systems Open
Cloud storage services offer a scalable platform to store a large amount of data at a low cost. It attracts a large number of customers to outsource their data to the cloud. To manage the massive growth in the size of outsourced data, clou…
View article: Clinical QA 2.0: Multi-Task Learning for Answer Extraction and Categorization
Clinical QA 2.0: Multi-Task Learning for Answer Extraction and Categorization Open
Clinical Question Answering (CQA) plays a crucial role in medical decision-making, enabling physicians to extract relevant information from Electronic Medical Records (EMRs). While transformer-based models such as BERT, BioBERT, and Clinic…
View article: Predicting breast cancer recurrence using deep learning
Predicting breast cancer recurrence using deep learning Open
Breast cancer and its recurrence are significant health concerns, emphasizing the critical importance of early detection and personalized treatment strategies for improved outcomes. This study introduces the BCR-HDL (Breast Cancer Recurren…
View article: MVTamperBench: Evaluating Robustness of Vision-Language Models
MVTamperBench: Evaluating Robustness of Vision-Language Models Open
Multimodal Large Language Models (MLLMs), are recent advancement of Vision-Language Models (VLMs) that have driven major advances in video understanding. However, their vulnerability to adversarial tampering and manipulations remains under…
View article: Survey of Large Multimodal Model Datasets, Application Categories and Taxonomy
Survey of Large Multimodal Model Datasets, Application Categories and Taxonomy Open
Multimodal learning, a rapidly evolving field in artificial intelligence, seeks to construct more versatile and robust systems by integrating and analyzing diverse types of data, including text, images, audio, and video. Inspired by the hu…
View article: Enhancing Document AI Data Generation Through Graph-Based Synthetic\n Layouts
Enhancing Document AI Data Generation Through Graph-Based Synthetic\n Layouts Open
The development of robust Document AI models has been constrained by limited\naccess to high-quality, labeled datasets, primarily due to data privacy\nconcerns, scarcity, and the high cost of manual annotation. Traditional methods\nof synt…
View article: Production of electricity from urine
Production of electricity from urine Open
The research work explores the possibility of utilizing urine, the most abundant waste on earth as an unconventional, yet, plausible alternative to generate electricity. A groundbreaking two-phase method has been introduced that utilizes a…
View article: A Systematic Survey of Multi-Factor Authentication for Cloud Infrastructure
A Systematic Survey of Multi-Factor Authentication for Cloud Infrastructure Open
The unauthorized usage of various services and resources in cloud computing is something that must be protected against. Authentication and access control are the most significant concerns in cloud computing. Several researchers in this fi…
View article: A Slice-Based Change Impact Analysis for Regression Test Case Prioritization of Object-Oriented Programs
A Slice-Based Change Impact Analysis for Regression Test Case Prioritization of Object-Oriented Programs Open
Test case prioritization focuses on finding a suitable order of execution of the test cases in a test suite to meet some performance goals like detecting faults early. It is likely that some test cases execute the program parts that are mo…