Publications

^* Equal contribution

2026

1. XRPO: Pushing the Limits of GRPO with Targeted Exploration and Exploitation

   Udbhav Bamba^*, Minghao Fang^*, Yifan Yu^*, and 2 more authors

   In International Conference on Machine Learning (ICML), 2026

   Bib PDF

   @inproceedings{bamba2025xrpo,
     title = {XRPO: Pushing the Limits of GRPO with Targeted Exploration and Exploitation},
     author = {Bamba, Udbhav and Fang, Minghao and Yu, Yifan and Zheng, Haizhong and Lai, Fan},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2026},
     url = {https://arxiv.org/abs/2510.06672},
   }

2. DOT-MoE: Differentiable Optimal Transport for MoEfication

   Udbhav Bamba^*, Arnav Chavan^*, Aryamaan Thakur^*, and 2 more authors

   In International Conference on Machine Learning (ICML), 2026

   Bib PDF

   @inproceedings{bamba2026dotmoe,
     title = {DOT-MoE: Differentiable Optimal Transport for MoEfication},
     author = {Bamba, Udbhav and Chavan, Arnav and Thakur, Aryamaan and Teig, Steve and Gupta, Deepak K.},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2026},
     url = {https://arxiv.org/abs/2606.01666},
   }

3. S2D: Selective Spectral Decay for Quantization-Friendly Conditioning of Neural Activations

   Arnav Chavan^*, Nahush Lele^*, Udbhav Bamba^*, and 3 more authors

   In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2026

   Bib PDF

   @inproceedings{chavan2025s2d,
     title = {S2D: Selective Spectral Decay for Quantization-Friendly Conditioning of Neural Activations},
     author = {Chavan, Arnav and Lele, Nahush and Bamba, Udbhav and Dayal, Sankalp and Raghunathan, Aditi and Gupta, Deepak K.},
     booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
     year = {2026},
     url = {https://arxiv.org/abs/2602.14432},
   }

4. CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models

   Neeraj Anand, Samyak Jha, Udbhav Bamba, and 1 more author

   Transactions on Machine Learning Research, 2026

   Bib PDF

   @article{anand2025crops,
     title = {CRoPS: A Training-Free Hallucination Mitigation Framework for Vision-Language Models},
     author = {Anand, Neeraj and Jha, Samyak and Bamba, Udbhav and Rahaman, Rahul},
     journal = {Transactions on Machine Learning Research},
     year = {2026},
     url = {https://openreview.net/forum?id=KQSoZDPVGX},
   }

5. Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models

   Rishabh Tiwari^*, Aditya Tomar^*, Udbhav Bamba^*, and 5 more authors

   In International Conference on Machine Learning (ICML), 2026

   Bib PDF

   @inproceedings{bamba2026reward,
     title = {Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models},
     author = {Tiwari, Rishabh and Tomar, Aditya and Bamba, Udbhav and Maheswaran, Monishwaran and Yang, Heng and Mahoney, Michael W. and Keutzer, Kurt and Gholami, Amir},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2026},
     url = {https://arxiv.org/abs/2603.06621},
   }

2025

1. Reward Under Attack: Evaluating the Sensitivity of Process Reward Models

   Udbhav Bamba^*, Heng Yang^*, Rishabh Tiwari^*, and 2 more authors

   In 2nd AI for Math Workshop at ICML, 2025

   Bib PDF

   @inproceedings{bamba2025reward,
     title = {Reward Under Attack: Evaluating the Sensitivity of Process Reward Models},
     author = {Bamba, Udbhav and Yang, Heng and Tiwari, Rishabh and Keutzer, Kurt and Gholami, Amir},
     booktitle = {2nd AI for Math Workshop at ICML},
     year = {2025},
     url = {https://openreview.net/forum?id=Hw24VOppus},
   }

2024

1. Partial Binarization of Neural Networks for Budget-Aware Efficient Learning

   Udbhav Bamba^*, Neeraj Anand^*, Saksham Aggarwal^*, and 2 more authors

   In IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2024

   Bib PDF

   @inproceedings{Bamba_2024_WACV,
     author = {Bamba, Udbhav and Anand, Neeraj and Aggarwal, Saksham and Prasad, Dilip K. and Gupta, Deepak K.},
     title = {Partial Binarization of Neural Networks for Budget-Aware Efficient Learning},
     booktitle = {IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
     year = {2024},
     url = {https://ieeexplore.ieee.org/document/10483283},
   }

2. An UltraMNIST classification benchmark to train CNNs for very large images

   Udbhav Bamba^*, Deepak K. Gupta^*, Abhishek Thakur^*, and 6 more authors

   In Scientific Data, 2024

   Bib PDF

   @inproceedings{Gupta2024UltraMNIST,
     title = {An UltraMNIST classification benchmark to train CNNs for very large images},
     author = {Bamba, Udbhav and Gupta, Deepak K. and Thakur, Abhishek and Gupta, Akash and Agarwal, Rohit and Sharan, Suraj and Demir, Ertugul and Agarwal, Krishna and Prasad, Dilip K.},
     booktitle = {Scientific Data},
     year = {2024},
     url = {https://doi.org/10.1038/s41597-024-03587-4},
   }

2022

1. Finding Structured Winning Tickets with Early Pruning

   Udbhav Bamba, Devin Kwok, Gintare Karolina Dziugaite, and 1 more author

   In Hardware Aware Efficient Training (HAET) Workshop at ICML, 2022

   Bib PDF

   @inproceedings{bamba2022kernalpruning,
     title = {Finding Structured Winning Tickets with Early Pruning},
     author = {Bamba, Udbhav and Kwok, Devin and Karolina Dziugaite, Gintare and Rolnick, David},
     booktitle = {Hardware Aware Efficient Training (HAET) Workshop at ICML},
     year = {2022},
     url = {https://haet2022.github.io/accepter_papers},
   }

2. Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning

   Arnav Chavan^*, Rishabh Tiwari^*, Udbhav Bamba^*, and 1 more author

   In IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022

   Bib PDF

   @inproceedings{Chavan2021MetaDock,
     author = {Chavan, Arnav and Tiwari, Rishabh and Bamba, Udbhav and Gupta, Deepak K.},
     booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
     title = {Dynamic Kernel Selection for Improved Generalization and Memory Efficiency in Meta-learning},
     year = {2022},
     url = {https://ieeexplore.ieee.org/document/9878632},
   }

2021

1. ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations

   Rishabh Tiwari^*, Udbhav Bamba^*, Arnav Chavan^*, and 1 more author

   In International Conference on Learning Representations (ICLR), 2021

   Bib PDF

   @inproceedings{tiwari2021chipnet,
     title = {ChipNet: Budget-Aware Pruning with Heaviside Continuous Approximations},
     author = {Tiwari, Rishabh and Bamba, Udbhav and Chavan, Arnav and Gupta, Deepak K.},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2021},
     url = {https://openreview.net/forum?id=xCxXwTzx4L1},
   }

2. Rescaling CNN Through Learnable Repetition of Network Parameters

   Udbhav Bamba^*, Arnav Chavan^*, Rishabh Tiwari^*, and 1 more author

   In IEEE International Conference on Image Processing (ICIP), 2021

   Bib PDF

   @inproceedings{Chavan2021RepeatNet,
     author = {Bamba, Udbhav and Chavan, Arnav and Tiwari, Rishabh and Gupta, Deepak},
     booktitle = {IEEE International Conference on Image Processing (ICIP)},
     title = {Rescaling CNN Through Learnable Repetition of Network Parameters},
     year = {2021},
     url = {https://ieeexplore.ieee.org/document/9506158},
   }