Addison J. Wu
I'm a senior at Princeton University studying computer science and math
I want to build AI systems that interact better with each other, with us, and with the world over time, and ensure they help us do even greater things we find worth doing. Check out what I've been working on!
2024 – now
-
Multi-agent and human-AI interaction
Large Language Models Develop Novel Social Biases Through Adaptive Exploration
Addison J. Wu*, Ryan Liu*, Xuechunzi Bai, Thomas L. Griffiths
ICML 2026 Oral [arXiv]
Media coverage: The Guardian, Forbes, MIT Technology Review, Hacker News (Y Combinator), The Times of India, Gizmodo, Arizona's Family
How does Adversarial Influence Scale in Multi-Agent Systems?
Addison J. Wu*, Jasin Cekinmez*, Michel Liao*, Karthik Narasimhan, Thomas L. Griffiths
arXiv preprint, 2026 [arXiv]
Ads in AI Chatbots? An Analysis of How Large Language Models Navigate Conflicts of Interest
Addison J. Wu*, Ryan Liu*, Shuyue Stella Li, Yulia Tsvetkov, Thomas L. Griffiths
COLM 2026 [arXiv]
Media coverage: Hacks/Hackers, The Indian Express; Cited in Senate testimony
Are Large Language Models Sensitive to the Motives Behind Communication?
Addison J. Wu*, Ryan Liu*, Kerem Oktar*, Theodore R. Sumers, Thomas L. Griffiths
NeurIPS 2025; PragLM @ COLM 2025 Oral [arXiv]
Query Timing Produces Opposite Positional Biases Between LLMs and Humans
Jasin Cekinmez*, Addison J. Wu*, Thomas L. Griffiths
EMNLP 2026 Findings; ICBINB @ ICLR 2026 Entropic Paper Award [arXiv]
Mind Your Step (by Step): Chain-of-Thought can Reduce Performance on Tasks where Thinking Makes Humans Worse
Ryan Liu*, Jiayi Geng*, Addison J. Wu, Ilia Sucholutsky, Tania Lombrozo, Thomas L. Griffiths
ICML 2025 [arXiv]
Media coverage: The Information [text] -
Data and training constraints
Quantifying Empirical Compute-Supervision Tradeoffs in RLVR
Ryo Mitsuhashi*, Patrick Chen*, Isabelle Tseng*, Jasin Cekinmez*, Addison J. Wu*
Workshop on Combining Benchmarks and Theory @ ICML 2026 [arXiv]
Where did the ambiguity go? Examining how multimodal models interpret polysemous words
Jasin Cekinmez*, Addison J. Wu*, Raja Marjieh, Thomas L. Griffiths
Sci-FM @ COLM 2026 Oral [arXiv]
Guess the Unified Model: How Much Can We Recover from Generated Images?
Jasin Cekinmez*, Ryo Mitsuhashi*, Addison J. Wu*, Yida Yin
under review, 2026 [arXiv] -
Visual reasoning about the physical world
SportD: Can VLMs Physically Strategize?
Jasin Cekinmez*, Addison J. Wu*, …, Weining Shen
PhysUnderstand @ NeurIPS 2026 Oral [arXiv]
OccludeBench: Can VLMs Find What's Missing?
Jasin Cekinmez*, Addison J. Wu*, Ryo Mitsuhashi*, Linrong Cai, Xingyu Fu, Zhuang Liu
Computer Vision in the Wild @ CVPR 2026 [arXiv]
2026 – now
with Mantas Mazeika and
Dan Hendrycks
-
Reward hacking, other projects coming soon
2023 – 2025
Multi-Physics Interaction Lab, University of Waterloo
with Jean-Pierre Hickey
-
Toward a Neural Network-Based Approach for Improved Atmospheric Infrasound Source Localization
Addison J. Wu, Arnav Joshi, Jean-Pierre Hickey
IEEE Access, 2025 [Paper]
I love to travel ✈️
🇨🇦 🇺🇸 🇲🇽 🇵🇷 🇨🇺 🇵🇦 🇧🇷 🇦🇷 🇯🇵 🇰🇷 🇨🇳 🇭🇰 🇮🇪 🇬🇧 🇫🇷 🇩🇪 🇪🇸 🇧🇪 🇨🇭 🇮🇹 🇻🇦 🇸🇲 🇩🇰 🇸🇪 🇳🇴