Email  ·  LinkedIn  ·  Scholar  ·  GitHub

About

Hi, I’m Pratham, an M.S. student in Computer Science at UC San Diego. I work in the Spatiotemporal ML Lab and am advised by Prof. Rose Yu.

Research

My research interests include multimodal learning, time series modeling, and interpretability in multimodal models. I am especially interested in how models can align and reason across text, vision, and numeric time series data.

Recently, I have worked on multimodal time series benchmarks, contrastive trimodal alignment (time series, vision, and language), artifact propogation in captioning-inpainting frameworks and ML for biomechanical sensor data. I am currently exploring topics in physical reasoning in multimodal models and multimodal steering.

Publications and Preprints

Time Series, Vision, and Language: Exploring the Limits of Alignment in Contrastive Representation Spaces
Pratham Yashwante, Rose Yu
ICML 2026  ·  arXiv

How Do Inpainting Artifacts Propagate to Language?
Pratham Yashwante, Davit Abrahamyan, Shresth Grover, Sukruth Rao
ACL Main 2026  ·  arXiv

CaTS-Bench: Can Language Models Describe Time Series?
Luca Zhou*, Pratham Yashwante*, Marshall Fisher, Alessio Sampieri, Zihao Zhou, Fabio Galasso, Rose Yu
ACL Findings 2026  ·  arXiv

A Multi-Sensor, Multi-Movement Study of Motion Tape Strain Data for Low Back Pain Classification
Pratham Yashwante, Sara P. Gombatto, Yasmín Velázquez, Elijah Wyckoff, Aarti Lalwani, Kevin Patrick, Kenneth Loh, Emilia Farcas, Rose Yu
Under review at Sensors, 2026  ·  arXiv coming soon

A Hybrid Deep Learning Model for Automated Cricket Commentary Generation
Pratham Yashwante, Yash Patil, Bhavika Wani, Suvarna Chaure
IEEE ICCMLA 2024  ·  IEEE Xplore

A Hybrid Sentiment and Emotion Analysis Model for Marathi Text Using Horse Herd Optimization, Bidirectional RNN, and Affective Cognitive Computing
Anindita Khade, Pratham Yashwante, Devesh Shetty
Cognitive Computing and Cyber Physical Systems, 2024  ·  Springer Nature

Comparative Analysis of Meta-heuristic Feature Selection and Feature Extraction Approaches for Enhanced Chronic Kidney Disease Prediction
Pratham Yashwante, Yash Patil, Karan Nadar, Anindita Khade
IEEE IATMSI 2024  ·  IEEE Xplore

Education

University of California, San Diego
M.S. in Computer Science  ·  Sep 2024 – Jun 2026 (expected)  ·  GPA: 4.0/4.0

University of Mumbai, SIES Graduate School of Technology
B.E. in Computer Engineering  ·  Jan 2021 – Jul 2024  ·  GPA: 9.47/10

Activities

Reviewer — CHIL 2026, AAAI 2025