Part of my facial expressions series. All of these LoRAs are designed to mesh well with each other and have low impact on your image framing.
Straightforward, but a little subtle. It's trained low to work well with other aspects of expressions (some examples of mixes in the gallery), I recommend increasing the weight of pleading in the prompt over increasing the LoRA weight.
Newly updated to reduce the default-horniness and framing/context bias like arms-over-head positioning. Am very happy with how the adjustments turned out, will be taking a pass with some of my other LoRAs to do the same, so let me know if there are other biases you're still seeing.
Recommended starting prompts:
<subject>, pleading eyes
<subject>, pleading expression