Rizwan Khan

[Rahul Chhabra @rahulchhabra07]

Introducing RL with Neural Feedback (RLNF). We have found reliable ways to train AI models to respond to human preferences purely from brain signals.

The example below is a picture of a face that self-transforms into a more attractive version for the subject as they are looking at it.

[Rizwan Khan @rizwanbhatii122]

@zainabkhan673 Impressive execution

[Rizwan Khan @rizwanbhatii122]

@Humabaloxh This stands out

[Rizwan Khan @rizwanbhatii122]

@saravibes1714 Quality keeps improving

[Rizwan Khan @rizwanbhatii122]

@Sapnahohr Quality keeps improving

[Rizwan Khan @rizwanbhatii122]

@sufia_khan23 AI keeps moving fast

[Rizwan Khan @rizwanbhatii122]

@Kaleemkalei7 Looking forward to trying it

[Rizwan Khan @rizwanbhatii122]

Nice breakdown

[Rizwan Khan @rizwanbhatii122]

Instead of thinking of an AI project as a sequence of disconnected tasks, imagine keeping the entire creative process moving through one conversation. That's the idea behind Flova 1.5's multi-Skill workflow. You can tell the Agent what you want to achieve, let it plan the process, and continue making changes as the project develops. https://t.co/gW22B4Z9qr

[FlovaAI @Flovaai]

Flova 1.5 | One Chat with Multiple Skills
One project. Multiple Skills.
Bring different Skills together in one creation!

- You set the goal. Agent plans the workflow.
- Mix and match Skills to explore different stories and styles.
- Refine your creations and take your ideas even further.

✨ Mix and Match Skills. Turn your ideas into videos, your way!

[Rizwan Khan @rizwanbhatii122]

@Fatima_noor88 Smart approach

[Rizwan Khan @rizwanbhatii122]

@laiba_malik3 The pace of AI is incredible

[Rizwan Khan @rizwanbhatii122]

@rauf1714 This changes a lot

[Rizwan Khan @rizwanbhatii122]

@Flovaai The most interesting part is how much freedom you get to shape the result along the way.

[FlovaAI @Flovaai]

Flova 1.5 | One Chat with Multiple Skills
One project. Multiple Skills.
Bring different Skills together in one creation!

- You set the goal. Agent plans the workflow.
- Mix and match Skills to explore different stories and styles.
- Refine your creations and take your ideas even further.

✨ Mix and Match Skills. Turn your ideas into videos, your way!

[Rizwan Khan @rizwanbhatii122]

@rauf1714 AI keeps moving fast

[Rizwan Khan @rizwanbhatii122]

@laiba_malik3 Quality keeps improving

[Rizwan Khan @rizwanbhatii122]

@Fatima_noor88 This stands out

[Rizwan Khan @rizwanbhatii122]

Impressive execution

[Rizwan Khan @rizwanbhatii122]

After this thread I'm less interested in new voice models and more interested in what they're trained on.

[Shobhit Banga @shobhitbanga]

Drop 1/15: Introducing Joy, a dual-channel conversational dataset.
Real, unscripted conversations between two people, each on their own track. Built to teach voice models how people actually talk: when to listen, when to jump in, when to say “hmm”.

100,000 hours.
50,000 speakers.
10 languages.

[Rizwan Khan @rizwanbhatii122]

@dylancooper_ai Impressive executionGreat progress

[Rizwan Khan @rizwanbhatii122]

@Sapnahohr Impressive execution