Abhinav Agarwalla

MagiQ Technologies (United States) — United States

Computer Science · Physical Sciences

5h-index1.5kcitations18works0.02yr avg
OpenAlex↗ORCID↗Google Scholar↗Lab page↗

Accepting Students?

No reports yet. Know if this professor is taking students?

Research Topics

Video Surveillance and Tracking Methods(5), Robotic Mechanisms and Dynamics(4), Multimodal Machine Learning Applications(4), Domain Adaptation and Few-Shot Learning(3), Anomaly Detection Techniques and Applications(3)

Publications18 total

Sparse Fine-Tuning for Inference Acceleration of Large Language Models
Machine translation·2025
Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment
arXiv (Cornell University)·2024Open Access
Lidar Panoptic Segmentation and Tracking without Bells and Whistles
arXiv (Cornell University)·2023Open Access
Lidar Panoptic Segmentation and Tracking without Bells and Whistles
2023· 6 cited
Dense or Sparse: Crowd Counting with Binary Supervision
2022
Beyond Learning Features: Training a Fully-Functional Classifier with ZERO Instance-Level Labels
Proceedings of the AAAI Conference on Artificial Intelligence·2022· 1 citedOpen Access
Completely Self-supervised Crowd Counting via Distribution Matching
Lecture notes in computer science·2022· 22 cited
Completely Self-Supervised Crowd Counting via Distribution Matching
arXiv (Cornell University)·2020· 3 citedOpen Access
Unsupervised Domain Adaptation for Learning Eye Gaze from a Million Synthetic Images
2018· 8 cited
Unsupervised Domain Adaptation for Learning Eye Gaze from a Million Synthetic Images: An Adversarial Approach
arXiv (Cornell University)·2018· 3 citedOpen Access
Unsupervised Domain Adaptation for Learning Eye Gaze from a Million\n Synthetic Images: An Adversarial Approach
arXiv (Cornell University)·2018· 1 citedOpen Access
Bayesian optimisation with prior reuse for motion planning in robot soccer
2018Open Access
Representation-Aggregation Networks for Segmentation of Multi-Gigapixel Histology Images
arXiv (Cornell University)·2017· 18 citedOpen Access
Recurrent Memory Addressing for Describing Videos
2017· 2 citedOpen Access
NTIRE 2017 Challenge on Single Image Super-Resolution: Methods and Results
2017· 1.5k cited
Hierarchical Ranking of Cricket Teams Incorporating Player Composition
Lecture notes in computer science·2017· 2 cited
Bayesian Optimisation with Prior Reuse for Motion Planning in Robot\n Soccer
arXiv (Cornell University)·2016Open Access
Optimal Spline Trajectories by Modelling Kinematic Constraints in Robot Soccer.
arXiv (Cornell University)·2016Open Access

Frequent Co-authors

Arnav Jain(5), Deepak Babu Sam(4), Jimmy Joseph(3), Avisek Lahiri(3), Prabir Kumar Biswas(3), Jayanta Mukhopadhyay(3), Shuicheng Yan(2), Vishwanath A. Sindagi(2), R. Venkatesh Babu(2), Vishal M. Patel(2), Xuhua Huang(2), Jason Ziglar(2), Francesco Ferroni(2), Laura Leal-Taixé(2), James Hays(2), Aljoša Ošep(2), Deva Ramanan(2), Venkatesh Babu Radhakrishnan(2), KV Manohar(2), Arpit Saxena(2)