Abhishek Kumar

Research Scientist, Google DeepMind

Joined

September 2016

Names

Abhishek Kumar (Preferred)

Emails

****@umiacs.umd.edu (Confirmed)

****@us.ibm.com (Confirmed)

****@inductivebias.net (Confirmed)

****@gmail.com (Confirmed)

****@google.com (Confirmed)

Personal Links

Homepage

Google Scholar

DBLP

Career & Education History

Research Scientist

Google DeepMind (google.com)

2018 – Present

Research Staff Member

IBM Research, Yorktown Heights, NY (us.ibm.com)

2013 – 2018

PhD student

University of Maryland College Park (umiacs.umd.edu)

2010 – 2013

Advisors, Relations & Conflicts

Coauthor

Piyush Rai

Present

PhD Advisor

Hal Daume III

2010 – 2013

Expertise

Machine learning

deep learning

representation learning

generative models

distribution shift

label noise

transfer learning

Present

Publications

Guiding Generative Recommender Systems with Structured Human Priors via Multi-head Decoding
Yunkai Zhang, Qiang Zhang, Diji Yang, Ryan Feng Lin, Ruizhong Qiu, Benyu Zhang, Hanchao Yu, Jiayi Liu, Yinglong Xia, Zhuokai Zhao, Lizhu Zhang, Xiangjun Fan, Zhuoran Yu, Abhishek Kumar, Zeyu Zheng
- The Web Conference (WWW) 2026
- Readers: Everyone
Wanda++: Pruning Large Language Models via Regional Gradients
Yifan Yang, Kai Zhen, Bhavana Ganesh, Aram Galstyan, Goeric Huybrechts, Markus Müller, Jonas M. Kübler, Rupak Vignesh Swaminathan, Athanasios Mouchtaris, Sravan Babu Bodapati, Nathan Susanj, Zheng Zhang, Jack FitzGerald, Abhishek Kumar
- SLLM
- Readers: Everyone
RB-Modulation: Training-Free Stylization using Reference-Based Modulation
Litu Rout, Yujia Chen, Nataniel Ruiz, Abhishek Kumar, Constantine Caramanis, Sanjay Shakkottai, Wen-Sheng Chu
- ICLR 2025 Oral
- Readers: Everyone
The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
Tyler LaBonte, John Collins Hill, Xinchen zhang, Vidya Muthukumar, Abhishek Kumar
- NeurIPS 2024 poster
- Readers: Everyone
Early Weight Averaging meets High Learning Rates for LLM Pre-training
Sunny Sanyal, Atula Tejaswi Neerkaje, Jean Kaddour, Abhishek Kumar, sujay sanghavi
- COLM
- Readers: Everyone
Beyond Human Data: Scaling Self-Training for Problem-Solving with Language Models
Avi Singh, John D Co-Reyes, Rishabh Agarwal, Ankesh Anand, Piyush Patil, Xavier Garcia, Peter J Liu, James Harrison, Jaehoon Lee, Kelvin Xu, Aaron T Parisi, Abhishek Kumar, Alexander A Alemi, Alex Rizkowsky, Azade Nova, Ben Adlam, Bernd Bohnet, Gamaleldin Fathy Elsayed, Hanie Sedghi, Igor Mordatch et al. (20 additional authors not shown)
- Accepted by TMLR
- Readers: Everyone
Early Weight Averaging meets High Learning Rates for LLM Pre-training
Sunny Sanyal, Atula Tejaswi Neerkaje, Jean Kaddour, Abhishek Kumar, sujay sanghavi
- WANT@NeurIPS 2023 Poster
- Readers: Everyone
Early Weight Averaging Meets High Learning Rates for LLM Pre-training
Sunny Sanyal, Atula Tejaswi Neerkaje, Jean Kaddour, Abhishek Kumar, sujay sanghavi
- ICLR 2024 Conference Withdrawn Submission
- Readers: Everyone
Towards Last-layer Retraining for Group Robustness with Fewer Annotations
Tyler LaBonte, Vidya Muthukumar, Abhishek Kumar
- 22 Sept 2023
- OpenReview Archive Direct Upload
- Readers: Everyone
Small-scale proxies for large-scale Transformer training instabilities
Mitchell Wortsman, Peter J Liu, Lechao Xiao, Katie E Everett, Alexander A Alemi, Ben Adlam, John D Co-Reyes, Izzeddin Gur, Abhishek Kumar, Roman Novak, Jeffrey Pennington, Jascha Sohl-Dickstein, Kelvin Xu, Jaehoon Lee, Justin Gilmer, Simon Kornblith
- ICLR 2024 oral
- Readers: Everyone

View all 49 publications

Co-Authors

View all 155 co-authors