Hi, I notice that given the same input keywords, across different runs, the generated text are the same, even setting different seeds by 'pl.seed_everything(..)'.

[RuiFeiHe]

Is your feature request related to a problem? Please describe.
A clear and concise description of what the problem is. Ex. I'm always frustrated when [...]

Describe the solution you'd like
A clear and concise description of what you want to happen.

Describe alternatives you've considered
A clear and concise description of any alternative solutions or features you've considered.

Additional context
Add any other context or screenshots about the feature request here.

[RuiFeiHe]

Hi, I notice that given the same input keywords, across different runs, the generated text are the same, even setting different seeds by 'pl.seed_everything(..)'.

Is there any way to generate different sentences given the same keywords?

[Ikkor]

Want to know this as well.

[ArchaicAccount]

I think you have to add a new word to the list, though if there is a different way that doesn't require changing the list I'd love to know aswell.

[akuma308]

@gagan3012 Please let us know if that cane be done .

[gagan3012]

Hello,
The models have been trained in such a way that there are only one sentence for one set of keywords. That is why the same inputs generate the same outputs

[netforcelabs1]

@gagan3012 Anyway it can return complete sentense minimum 2-3 lines for a particular keyword? is that possible?