HOrIc&RRO7HrEs#uNl NB%6GDjxO2Y0
My name is Justin Deschenaux and I am a PhD student at . I am advised by Professor Caglar Gulcehre, and I work on diffusion language models, a blazingly fast, controllable, and principled way to generate text.
Latest News
- Nov. 2025
We started a reading group on discrete diffusion with Subham and Zhihan.
- May. 2025
The Diffusion Duality was accepted at ICML.
- Jan. 2025
SDTT was accepted at ICLR.
- Sep 2024
Started PhD at Caglar Gulcehre.
Selected Work
1
Partition Generative Modeling: Masked Modeling Without Masks
PGMs partition tokens to parallelize decoding with massive latency gains.
# Generative Models# Masked Modeling# Sparse Attention# Efficiency
2
Loopholing Discrete Diffusion: Deterministic Bypass of the Sampling Wall
Loopholing keeps diffusion latents informative, slashing perplexity and boosting reasoning.
# Diffusion Models# Language Modeling# Latent Variables# Reasoning