Pura Peetathawatchai
@poonpura.bsky.social
M.S. Computer Science @Stanford. Interested in machine learning privacy, AI security, diffusion models, cryptography, AI for environment, healthcare, education
🌱 poonpura.github.io
🌱 poonpura.github.io
Wise words from a fellow scholar…
November 20, 2025 at 2:57 PM
Wise words from a fellow scholar…
We tried generating images using different values of subsample size (m) and DP parameter ε. Our results were particularly good for Textual Inversion (TI)!
🧵 15/16
🧵 15/16
November 27, 2024 at 6:43 PM
We tried generating images using different values of subsample size (m) and DP parameter ε. Our results were particularly good for Textual Inversion (TI)!
🧵 15/16
🧵 15/16
We tested the effectiveness of our approach on two different target datasets: a collection of artworks from an artist (with consent, see her art on Instagram: @eveismyname) and the Paris 2024 Olympic pictograms (approved for non-commercial editorial use, ©️IOC - 2023)
🧵 14/16
🧵 14/16
November 27, 2024 at 6:43 PM
We tested the effectiveness of our approach on two different target datasets: a collection of artworks from an artist (with consent, see her art on Instagram: @eveismyname) and the Paris 2024 Olympic pictograms (approved for non-commercial editorial use, ©️IOC - 2023)
🧵 14/16
🧵 14/16
How might we get a diffusion model to “learn” an art style without copying specific artworks? 🎨
🧵Let’s find out! (1/16)
📄: arxiv.org/abs/2411.14639
🧵Let’s find out! (1/16)
📄: arxiv.org/abs/2411.14639
November 27, 2024 at 6:43 PM
How might we get a diffusion model to “learn” an art style without copying specific artworks? 🎨
🧵Let’s find out! (1/16)
📄: arxiv.org/abs/2411.14639
🧵Let’s find out! (1/16)
📄: arxiv.org/abs/2411.14639