A CTR-driven advertising image generation pipeline that pre-trains an MLLM prompt model, trains an MLLM pairwise reward model on real click data, and fine-tunes with DPO plus a product-centric preference loss.
GPT-4V(ision) System Card
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CTR-Driven Advertising Image Generation with Multimodal Large Language Models
A CTR-driven advertising image generation pipeline that pre-trains an MLLM prompt model, trains an MLLM pairwise reward model on real click data, and fine-tunes with DPO plus a product-centric preference loss.