GRAINS Learning through connections

2026

• Paper titled "ModTGCN: Modularity-aware Graph Neural Networks for Text Classification" got accepted at the Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), June 09-12,2026 at Hong Kong in collaboration with Rajarshi Misra (UG researcher), Aditya Sharma, and Hari Om Aggrawal. A special congratulations to Rajarshi Misra for continuing this research even after completing his undergraduate studies at BITS Pilani — a true testament to his passion and commitment to the work. Congratulations to the entire team!

• We’re Bringing Graph Mining (CSF426) Back for CS Folks! Offered again in Semester II (2025-26) — One of BITS Pilani’s most sought-after and intellectually challenging electives! Not just popular — this course calls for strong effort, deep curiosity, and the motivation to uncover patterns from complex networks.📌 Handout and further details will be shared soon. Some past year feedbacks are available here.

• Poster accepted at Women in Machine Learning (WiML 2025) and travel grant of USD $1182.73 received for poster presentation at EurIPS 2025, Dec 01-08, 2025, in Copenhagen, Denmark. A big thanks to Nikita for co-organising the event and for creating opportunities for women worldwide to connect and network within the machine learning community.

• A paper titled "G-Loss: Graph-Guided Fine-Tuning of Language Models" got accepted at the Learning on Graphs (LoG) Conference [known for best quality reviews], December 10 - 12, 2025 in Arizona State University, Phoenix, USA with my PhD student Aditya Sharma and collaborator Prof. Rajesh Kumar from Bucknell University, USA. Congratulations, Aditya, for your continuous efforts and dedication! Your hard work finally paid off.  [paper]

• A paper titled "An Explainable Multimodal Framework with LLM Agents for Intracranial Hemorrhage Detection", has been accepted for presentation at the Agentic AI for Medicine Workshop @ MICCAI 2025, September 23-27 in Seoul, South Korea with Junior research fellow Shashwath Punneshetty, Undergraduate student Dhyey Italia (UG researcher) and collaborator Prof. Chandresh Maurya IIT Indore and Dr. Amit Agarwal AIIMS Bhopal. Congratulations Shashwath! and Best Wishes for new journey ahead at CMU.

• A paper titled "Weak Links in LinkedIn: Enhancing Fake Profile Detection in the Age of LLMs" was accepted at ASONAM with Apoorva (UG researcher), collaborator Prof. Rajesh Kumar (Bucknell University, USA), and my PhD student Aditya Sharma. This is the interesting research work led by Apoorva and Aditya, and it is a proud moment for me to have nurtured these two budding researchers. [webpage] [paper] [code] 

• We had a privilige of organizing a guest research talk by Raj Shah (PhD candidate at Georgia Institue of Technology, USA, Ex-BITS Pilani UG student) on topic "Using PLMs to understand Human Cognition" on September 8th, 2025.  A heartfelt thankyou to Raj for sharing his experience and taking time to interact with budding researchers at BITS Pilani.

• Delighted to host a research talk by Nikita Saxena (Research Engineer at Google Deepmind, Ex-BITSian UG researcher) on topic "Large Language Models: From pretraining to inference" on October 2nd, 2025.  Thankyou Nikita for a thought-provoking  and intellectually stimulating session.

• Guest talk by Aditya Sinha (Research Scientist at Netflix, Ex-BITSian UG researcher) on October 13, 2025. 

• “Adi Vaani” was featured in the Economic Times. This recognition is a proud moment for everyone involved.

• Our recent project “Adi Vaani”, sanctioned by the Ministry of Tribal Affairs and developed in collaboration with IIT Delhi, IIIT Hyderabad, IIIT Nava Raipur, and Tribal Research Institutes (TRIs), has successfully created a mobile application as well as a website. This project aims to bridge the gap between tribal languages of India and preserve low resource languages. Adi Vaani enables translation between four tribal languages — Mundari, Bhili, Santhali, and Gondi — and English/Hindi, supporting communication in both directions. BITS Pilani was in charge of the OCR development as well as building machine translation model for Mundari language.