Rhea Malhotra

I am a researcher at the Stanford Robotics Center and Interactive Robot Perception Lab, focusing on humanoids, mobile manipulation, and long-horizon robot learning. As a master's student at Stanford, I am advised by Jeannette Bohg and Oussama Khatib. I have been honored with the 2025 Firestone Medal for Research, awarded for the Best Honors Thesis in the Computer Science Department.

Email / Scholar / Github / CV / Art

Research

	HoMeR: Learning In-the-Wild Mobile Manipulation via Hybrid Imitation and Whole-Body Control Priya Sundaresan, Rhea Malhotra, Phillip Miao, Jingyun Yang, Jimmy Wu, Hengyuan Hu, Rika Antonova, Francis Engelmann, Dorsa Sadigh, Jeannette Bohg TBD, 2025 project page / arXiv We combines a hybrid IL agent with a fast, kinematics-based whole-body controller for sample-efficient, generalizable mobile manipulation in-the-wild.
	Humanoid Motion Scripting with Postural Synergies Rhea Malhotra, William Chong, Catie Cuan, Oussama Khatib Humanoids, 2025 (Oral Presentation) project page / arXiv We introduce SynSculptor, a humanoid motion analysis and editing framework that uses postural synergies to generate human-like motions for robots, leveraging motion capture data and a motion-language transformer to synthesize and evaluate new motions while adapting the robot's posture based on selected synergies.
	Self-Guided Action Diffusion Rhea Malhotra, Yuejiang Liu, Chelsea Finn RSS, 2025 (Oral Workshop) project page / arXiv We guide proposal action distribution at diffusion steps based on prior decisions. Under a tight sampling budget, our method achieves up to 70% higher success rates than existing counterparts on challenging dynamic tasks.
	CINner: modeling and simulation of chromosomal instability in cancer at single-cell resolution Khanh N Dinh, Ignacio Vázquez-García, Andrew Chan, Rhea Malhotra, Adam Weiner, Andrew W McPherson, Simon Tavaré PLOS Computational Biology, 2025 project page CINner is a versatile mathematical tool for modeling genomic changes in tumor evolution, including gene mutations, copy number alterations, and whole-genome duplication, to predict how chromosomal instability affects tumor growth and clonal heterogeneity across cancers.
	Single-cell multi-omics of mitochondrial DNA disorders reveals dynamics of purifying selection across human immune cells Caleb A Lareau, Sonia M Dubois, Frank A Buquicchio, Yu-Hsin Hsieh, Kopal Garg, Pauline Kautz, Lena Nitsch, Samantha D Praktiknjo, Patrick Maschmeyer, Jeffrey M Verboon, Jacob C Gutierrez, Yajie Yin, Evgenij Fiskin, Wendy Luo, Eleni P Mimitou, Christoph Muus, Rhea Malhotra, Sumit Parikh, Mark D Fleming, Lena Oevermann, Johannes Schulte, Cornelia Eckert, Anshul Kundaje, Peter Smibert, Santosha A Vardhana, Ansuman T Satpathy, Aviv Regev, Vijay G Sankaran, Suneet Agarwal, Leif S Ludwig Nature Genetics, 2023 project page Multi-omics approach to analyze pathogenic mitochondrial DNA (mtDNA) deletions and cell state features in single cells from six patients with single large-scale mtDNA deletions (SLSMDs), revealing distinct metabolic vulnerabilities and gene regulatory adaptations across cell types.
	Imaging of Drug Mules Rhea Malhotra, Ajay Singh Emergency Radiology, 2021 project page An overview of computational imaging techniques for detecting concealed drugs in drug mules, focusing on body packing, body pushing, and body stuffing. Key imaging enable computer vision detection with the use of abdominal radiography and low-dose CT to locate drug packages. Complications such as bowel obstruction and package rupture, underscoring the importance of radiological expertise in preventing harm and ensuring accurate diagnosis.

x

Credit to Priya Sundaresan for the HTML template!