Ava Pun

Ava Pun

Hi, my name is Ava! I’m a PhD student in the Computer Science Department of Carnegie Mellon University, advised by Jun-Yan Zhu. My research spans computer vision, computer graphics, and artificial intelligence. I wish to answer the question: How can we make AI useful not only on our computers, but also in the physical world?

Previously, I double-majored in Computer Science and Combinatorics & Optimization at the University of Waterloo. I also had the honour of interning at Waabi, where I conducted autonomous vehicle research under the supervision of Raquel Urtasun.

In my free time, I run the science webcomic Academy 118. I sometimes do photography.

Recent research

CubePart: An Open-Vocabulary Part-Controllable 3D Generator

arXiv preprint, 2026
Yiheng Zhu*, Kangle Deng*, Jean-Philippe Fauconnier*, Inaki Navarro*, Daiqing Li, Ava Pun, Yinan Zhang, Peiye Zhuang, Xiaoxia Sun, Maneesh Agrawala, Kiran Bhat, Tinghui Zhou
Interactive 3D assets used in games and simulation are typically decomposed into specific semantic parts to support animation, physics, and scripted behaviors, yet most generative 3D models produce either monolithic meshes or arbitrary part decompositions that cannot be aligned with application-specific requirements. We present CubePart, a generative framework for open-vocabulary, part-controllable 3D mesh generation that exposes part structure as an explicit inference-time control signal.…

Prompt-to-Product: Generative Assembly via Bimanual Manipulation

arXiv preprint, 2025
Ruixuan Liu*, Philip Huang*, Ava Pun, Kangle Deng, Shobhit Aggarwal, Kevin Tang, Michelle Liu, Deva Ramanan, Jun-Yan Zhu, Jiaoyang Li, Changliu Liu
Creating assembly products demands significant manual effort and expert knowledge in 1) designing the assembly and 2) constructing the product. This paper introduces Prompt-to-Product, an automated pipeline that generates real-world assembly products from natural language prompts. Specifically, we leverage LEGO bricks as the assembly platform and automate the process of creating brick assembly structures…

Generating Physically Stable and Buildable Brick Structures from Text

International Conference on Computer Vision (ICCV) 2025—Best Paper, Marr Prize
Ava Pun*, Kangle Deng*, Ruixuan Liu*, Deva Ramanan, Changliu Liu, Jun-Yan Zhu
We introduce BrickGPT, the first approach for generating physically stable interconnecting brick assembly models from text prompts. To achieve this, we construct a large-scale, physically stable dataset of brick structures, along with their associated captions, and train an autoregressive large language model to predict the next brick to add via next-token prediction. To improve the stability of the resulting designs, we employ an efficient validity check and physics-aware rollback during autoregressive inference…

Succinct Encodings of Binary Trees With Application to AVL Trees

Theoretical Computer Science, Volume 1056, 2025
Jeremy Chizewer*, Stephen Melczer*, J. Ian Munro*, Ava Pun*
We use a novel decomposition to create succinct data structures - supporting a wide range of operations on static trees in constant time - for a variety of tree classes, extending results of Munro, Nicholson, Benkner, and Wild. Motivated by the class of AVL trees, we further derive asymptotics for the information-theoretic lower bound on the number of bits needed to store tree classes whose generating functions satisfy certain functional equations…

Neural Lighting Simulation for Urban Scenes

Neural Information Processing Systems (NeurIPS) 2023
Ava Pun*, Gary Sun*, Jingkang Wang*, Yun Chen, Ze Yang, Sivabalan Manivasagam, Wei-Chiu Ma, Raquel Urtasun
Different outdoor illumination conditions drastically alter the appearance of urban scenes, and can harm the performance of image-based robot perception systems if they have not been seen during training. Towards this goal, we propose LightSim, a neural lighting camera simulation system that enables diverse, realistic, and controllable data generation. LightSim automatically builds relightable digital twins at scale from collected raw sensor data…

AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehicles

Conference on Computer Vision and Pattern Recognition (CVPR) 2021
Jingkang Wang, Ava Pun, James Tu, Sivabalan Manivasagam, Abbas Sadat, Sergio Casas, Mengye Ren, Raquel Urtasun
As self-driving systems become better, simulating scenarios where the autonomy stack may fail becomes more important. Traditionally, those scenarios are generated for a few scenes with respect to the planning module that takes ground-truth actor states as input. This does not scale and cannot identify all possible autonomy failures, such as perception failures due to occlusion. In this paper, we propose AdvSim, an adversarial framework to generate safety-critical scenarios…
AdvSim: Generating Safety-Critical Scenarios for Self-Driving Vehicles

*Denotes equal contribution or alphabetical order.

Latest blog posts

How to Build a Computer Using Photoshop Actions

A Turing machine implemented in Adobe Photoshop
  Ava Pun  •    Oct 14, 2023  •   10 min   computer-science  turing-machine  photoshop 
In this second instalment of Using Digital Art Programs for Unintended Purposes, I present you with a working computer programmed using Adobe Photoshop Actions. To be specific, this computer is a Turing machine—an abstract machine capable of running any computer algorithm. I’ll show you how to create this machine by using layer blend modes and the Invert command to perform logical computations.…

Conway’s Game of Life Implemented in Clip Studio Paint

  Ava Pun  •    Sep 15, 2023  •   9 min   computer-science  game-of-life  clip-studio-paint 
This is Conway’s Game of Life, implemented using the popular drawing software Clip Studio Paint (CSP). No, I’m not using the animation features in CSP, and I’m not drawing these pictures myself. Each frame is computed by Clip Studio Paint within the CSP file. The game advances every time you hit Save. How did I manage to squeeze this functionality out of a program that isn’t supposed to let you do anything more than edit images? Read on…