Summary of Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions Of Clean Data, by Jingyang Ou et al.
Your Absorbing Discrete Diffusion Secretly Models the Conditional Distributions of Clean Data
by Jingyang Ou, Shen Nie, Kaiwen Xue, Fengqi Zhu, Jiacheng Sun, Zhenguo Li, Chongxuan Li
First submitted to arxiv on: 6 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computation and Language (cs.CL)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper introduces a new type of discrete diffusion model called reparameterized absorbing discrete diffusion (RADD) that characterizes time-independent conditional probabilities. This model is designed to reduce the number of function evaluations by caching output when the noisy sample remains unchanged. The paper also shows that RADD can be unified with any-order autoregressive models, achieving state-of-the-art performance on zero-shot language modeling benchmarks at the GPT-2 scale. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This new type of diffusion model is simple and efficient, reducing the number of function evaluations by caching output when the noisy sample remains unchanged. The paper also shows that RADD can be used to unify absorbing discrete diffusion and any-order autoregressive models, achieving state-of-the-art performance on zero-shot language modeling benchmarks at the GPT-2 scale. |
Keywords
* Artificial intelligence * Autoregressive * Diffusion * Diffusion model * Gpt * Zero shot
