The latest LLMs such as Gemini Ultra and GPT-4o are capable of stunning feats of reasoning and natural language understanding. This is already powering a new wave of transformative AI use cases in e-commerce, and it appears that the pace of innovation is only just beginning to ramp up. E-commerce is undergoing a period of rapid change, with customers demanding seamless, personalized experiences across devices and channels. With vast product selections and dynamic consumer preferences, competition for customer attention and loyalty has become more intense than ever, while exciting new opportunities abound. Traditional approaches such as collaborative filtering or supervised learning may fall short in this complex, fast-paced world and fail to exhibit a deep understanding of customers’ concerns, feedback and evolving tastes.

The aim of this workshop is to foster discussion around the emerging role of LLMs in next-generation e-commerce applications, with a focus on topics that are relevant to the ICDM audience. The envisioned target audience will include both researchers in academia and industry practitioners who are interested in exploring the latest advances. We invite original and unpublished research contributions to LEMA in relevant subjects, including, but not limited to:

* Generative models for novel feature extraction from product descriptions
* Generating data driven, engaging product descriptions with LLMs
* Multimodal recommendations (combining text, image, and other data)
* Utilizing LLMs to process unstructured data and understand implicit user feedback
* LLM-enhanced user embeddings and personalization
* Optimizing voice search and multimodal data in e-commerce
* Fine-tuning vs. prompting - choosing the right approach for e-commerce LLMs
* Context-aware recommendations - leveraging user browsing history
* Integrating LLMs into large-scale recommendation workflows
* Bias and fairness in LLM-powered recommendation systems
* Ethical use of generative models in e-commerce