Jared Watrous

I received my BA in Computer Science and Minor in Math from UC Berkeley in May 2023, and my MS in Computer Science from Stanford in June 2025. I specialize in computer graphics/vision, 3D scene understanding, and related AI/ML applications.

I have worked as a Research Assistant at Stanford's NeuroAILab since October 2023, primarily focusing on 3D scene and dynamics understanding using Transformer-based deep learning. There I developed tools to collect multimodal datasets rivaling the scale of SOTA language data, and developed highly efficient inference frameworks for novel model architectures.

I have a strong background in low-level software development, systems design, and high-performance/parallel computing. I also have moderate experience with game and web development. In my free time, I like to create short animated films and write music.

Education

Stanford University - Computer Science MS	Sept 2023 - June 2025
UC Berkeley - Computer Science BA & Math Minor 1	Aug 2020 - May 2023
Pioneer High School - San Jose, CA 2	Class of 2020

¹Dean's List 2 terms, Distinction in General Scholarship
²Valedictorian, Class of 2020

Research

Stanford Neuroscience and Artificial Intelligence Laboratory (NeuroAILab)	Oct 2023 - Present
Research Assistant advised by Daniel Yamins Local Random Access Sequence Modeling, Probabilistic Structure Integration Transformer-based deep learning models for 3D scene understanding, dynamics, optical flow, depth, camera pose estimation, and image generation Large-scale data collection including procedurally generated 3D environments, automated scene rendering, and ethical web scraping of publicly available data, rivaling the scale of SOTA language datasets Design and implementation of high-performance inference frameworks for novel model architectures

Publications

• [arXiv 2025] World Modeling with Probabilistic Structure Integration
  Klemen Kotar, Wanhee Lee, Rahul Venkatesh, Honglin Chen, Daniel Bear, Jared Watrous, Simon Kim, Khai Loong Aw, Lilian Naing Chen, Stefan Stojanov, Kevin Feigelis, Imran Thobani, Alex Durango, Khaled Jedoui, Atlas Kazemian, Dan Yamins.
  Paper | https://doi.org/10.48550/arXiv.2509.09737
• [arXiv 2025] Discovering and using Spelke segments
  Rahul Venkatesh*, Klemen Kotar*, Lilian Naing Chen*, Seungwoo Kim, Luca Thomas Wheeler, Jared Watrous, Ashley Xu, Gia Ancone, Wanhee Lee, Honglin Chen, Daniel Bear, Stefan Stojanov, Daniel Yamins. *Equal contribution
  Website | Paper | https://doi.org/10.48550/arXiv.2507.16038
• [arXiv 2025] Taming generative video models for zero-shot optical flow extraction
  Seungwoo Kim*, Khai Loong Aw*, Klemen Kotar*, Cristobal Eyzaguirre, Wanhee Lee, Yunong Liu, Jared Watrous, Stefan Stojanov, Juan Carlos Niebles, Jiajun Wu, Daniel L. K. Yamins. *Equal contribution
  Website | Paper | https://doi.org/10.48550/arXiv.2507.09082
• [CCN 2025] On the Origin of 3D Perception in Visual World Models
  Wanhee Lee, Klemen Kotar, Jared Watrous, Rahul Mysore Venkatesh, Honglin Chen, Khai Loong Aw, Khaled Jedoui, Daniel LK Yamins. Cognitive Computational Neuroscience. University of Amsterdam, Netherlands, 2025.
  Extended Abstract
• [arXiv 2025] 3D Scene Understanding Through Local Random Access Sequence Modeling (LRAS)
  Wanhee Lee*, Klemen Kotar*, Rahul Mysore Venkatesh*, Jared Watrous*, Honglin Chen*, Khai Loong Aw, Daniel LK Yamins. *Equal contribution
  Website | Paper | https://doi.org/10.48550/arXiv.2504.03875
• [CCN 2024] A biologically plausible route to learn 3D perception
  Wanhee Lee*, Jared Watrous*, Honglin Chen, Klemen Kotar, Tyler Bonnen, Daniel Yamins. Cognitive Computational Neuroscience. Boston, Massachusetts, 2024. *Equal contribution
  Extended Abstract

Open Source Projects

• PyLRAS (WIP Fall 2025)
• Probabilistic Structure Integration
  Huggingface
• Pyramidal Local Patch Quantizer
  Huggingface (Source) | Huggingface (Checkpoint)

Academic Projects & Presentations

• [Stanford CS248A] Real-Time Soft Shadows
  Video Demo | Paper | Code
• [Stanford CS468] Spelke Object Segmentation with (Causal) Counterfactual World Modeling
  Slides (PDF) | Paper
• [Stanford CS224W] PyG PR: G-Mixup data augmentation for graph classification
  GitHub PR
• [Stanford CS348K] Light Culling Methods for Real-time Rendering Applications
  Slides (PDF) | Paper | Supplementary Video | Code
• [Stanford CS234] A Comparative Analysis of Reinforcement Learning Methods for Stratospheric Balloon Station Keeping
  Poster | Paper
• [Stanford CS448I] Exploring Unrolled Optimization
  Poster | Paper
• [Stanford CS236] Adapting Text-to-Image Models for Video Generation Using Few Shot Learning
  Poster | Paper
• [Stanford CS330] Combining Unsupervised Data Collection with Expert Demonstrations for Offline Multi-Task Robot Learning
  Poster | Paper
• [UC Berkeley LSC30Z] Droneup: A nature-inspired futuristic drone
  Poster

Animations

• 
  Laundry (Dec 2022 - Present)
  [WIP] 6min. independent short film inspired by Girls' Last Tour (Tsukumizu)
  Project Page
  
• 
  "Mr. Blue Sky" Salamander Dance (Apr 2020 - Jun 2020)
  90sec. dancing animation featuring Bruni from Frozen 2 (Disney)
  Project Page | YouTube
  
• 
  Willy Doesn't Have a UFO License (Mar 2024)
  20sec. demonstration of what happens when you give Willy (OC) the keys to your UFO
  Project Page | YouTube
  
• 
  Brewery Logo [Commission] (Mar 2022 - Apr 2023)
  Proprietary animated logo commissioned for a startup brewery
  Project Page
  
• 
  Pixar Intro (Apr 2020)
  My recreation of Pixar's classic intro sequence featuring Luxo Jr.
  Project Page
  
• See more 3D works  →

Last updated: 10/3/25