Telegram Group & Telegram Channel
tasty diffusion papers - september 2024

OmniGen: Unified Image Generation
what: one transformer for the text-to-image diffusion model.
- rectified flow.
- multimodal condition: text and image.
- one model processes all context and does diffusion steps.

Diffusion Policy Policy Optimization
what: set of best practices for fine-tuning diffusion-based policies in continuous control and robot learning tasks.

Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis
what: modification which helps to generalize better on few samples.

my thoughts
- i like how AI community is trying to simplify everything. surprisingly sometimes it works well. for example, rectified flow is simplified version of diffusion.
- rl + diffusion => next step in brain stimulation?
diffusion models could mimic brain patterns for smoother stimulation. with RL, they’d adapt in real-time, making treatments more precise and personalized. shifting from rigid protocols to dynamic brain interventions.

#digest



group-telegram.com/neural_cell/197
Create:
Last Update:

tasty diffusion papers - september 2024

OmniGen: Unified Image Generation
what: one transformer for the text-to-image diffusion model.
- rectified flow.
- multimodal condition: text and image.
- one model processes all context and does diffusion steps.

Diffusion Policy Policy Optimization
what: set of best practices for fine-tuning diffusion-based policies in continuous control and robot learning tasks.

Rejection Sampling IMLE: Designing Priors for Better Few-Shot Image Synthesis
what: modification which helps to generalize better on few samples.

my thoughts
- i like how AI community is trying to simplify everything. surprisingly sometimes it works well. for example, rectified flow is simplified version of diffusion.
- rl + diffusion => next step in brain stimulation?
diffusion models could mimic brain patterns for smoother stimulation. with RL, they’d adapt in real-time, making treatments more precise and personalized. shifting from rigid protocols to dynamic brain interventions.

#digest

BY the last neural cell




Share with your friend now:
group-telegram.com/neural_cell/197

View MORE
Open in Telegram


Telegram | DID YOU KNOW?

Date: |

'Wild West' Two days after Russia invaded Ukraine, an account on the Telegram messaging platform posing as President Volodymyr Zelenskiy urged his armed forces to surrender. On December 23rd, 2020, Pavel Durov posted to his channel that the company would need to start generating revenue. In early 2021, he added that any advertising on the platform would not use user data for targeting, and that it would be focused on “large one-to-many channels.” He pledged that ads would be “non-intrusive” and that most users would simply not notice any change. Soloviev also promoted the channel in a post he shared on his own Telegram, which has 580,000 followers. The post recommended his viewers subscribe to "War on Fakes" in a time of fake news. This ability to mix the public and the private, as well as the ability to use bots to engage with users has proved to be problematic. In early 2021, a database selling phone numbers pulled from Facebook was selling numbers for $20 per lookup. Similarly, security researchers found a network of deepfake bots on the platform that were generating images of people submitted by users to create non-consensual imagery, some of which involved children.
from ms


Telegram the last neural cell
FROM American