Improvements to prompt adherence. i.e., how closely the generation follows your prompt.
Note: This might mean some prompts from v1.0 might need to be slightly reworded. As v1.0 had a tendency to drop certain tokens. Sometimes, this resulted in unexpected but sometimes pleasant results. These results can still be achieved with a little bit of tweaking of the prompt.
Improvements to NSFW generation.
Likewise, the model should have less of a tendency to generate NSFW when not prompted to do such.
Additional training on Natural Language captions, using subset of NatViS dataset.
What's new in V2?
Improvements to prompt adherence. i.e., how closely the generation follows your prompt.
Note: This might mean some prompts from v1.0 might need to be slightly reworded. As v1.0 had a tendency to drop certain tokens. Sometimes, this resulted in unexpected but sometimes pleasant results. These results can still be achieved with a little bit of tweaking of the prompt.
Improvements to NSFW generation.
Likewise, the model should have less of a tendency to generate NSFW when not prompted to do such.
Additional training on Natural Language captions, using subset of NatViS dataset.
What's Next?
Continued Improvements to prompt adherence.
Inclusion of more Natural Language captions.