Bhawna Paliwal

About Me

Hi! I am Bhawna, Research Engineer at Microsoft Research (MSR) working with Dr. Manik Varma. My primary focus involves developing efficient language modeling architectures and training algorithms for applications in search and recommendation. I completed my undergraduate studies majoring in Computer Science at the Indian Institute of Technology (IIT), Ropar in 2021, where I was honored with the Director’s Gold Medal for my overall performance. My research interests lie in developing modular architectures for large-scale AI models and enabling equitable access to wider range of users by making them efficient and robust. If you are interested in machine learning research, engineering or solving impactful problems, I am sure we will have plenty to chat about :)

News

• [Oct. 2024] Our paper CROSS-JEM accepted at ENLSP@NeurIPS 2024!
• [May 2024] Our paper Extreme Meta-Classification for Large-Scale Zero-Shot Retrieval accepted at KDD 2024.
• [Mar. 2024] Our paper OAK: Enriching Document Representations using Auxiliary Knowledge for Extreme Classification accepted at ICML 2024.
• [Jan. 2024] Our paper Enhancing Tail Performance in Extreme Classifiers by Label Variance Reduction accepted at ICLR 2024.
• [Sep. 2023] Now Research Engineer II at MSR!
• [Jan. 2023] Presented my work on Real-time Query AutoCompletion at Microsoft Research India’s annual advisory board meeting.
• [Oct. 2022] Our paper Negative Mining-aware Mini-batching for Extreme Classification accepted at WSDM 2023.

Publications

Outreach

Presentations and Talks

• Got a chance to present our work on OAK (ICML'24) and CROSS-JEM (ENLSP@NeurIPS'24) to Bill Gates!!
  
• My talk on CROSS-JEM for ENLSP@NeurIPS 2024 available here
  
  Poster on CROSS-JEM
  

Reviewer

• Neural Information and Processing Systems (NeurIPS) 2024
• Knowledge Discovery and Data Mining (KDD) 2024

Volunteer

• Conference on Learning Theory (COLT) 2023
• WiCV, Computer Vision and Pattern Recognition (CVPR) 2021