Spotify uses Llama to deliver contextualized recommendations to boost artist discovery and create an even richer user experience. While Spotify’s users trust the audio streaming platform to connect them with content they’re likely to love, the streaming service discovered that people were four times more likely to engage with a creator’s content when given context about why it was recommended.
By combining Llama’s broad world knowledge and versatility with Spotify’s deep expertise in audio content, Spotify has created explanations that offer users personalized insights into their recommended content. The team has also created a way for its subscribers to receive personalized narratives about recommended new releases and culturally relevant commentary from their English and Spanish-speaking AI DJs. The result is an entertaining experience that feels tailor-made for each listener.
Compared to out-of-the-box Llama performance, Spotify discovered that domain-aware fine-tuning of Llama helped improve up to 14%. In a new blog post, the streaming service details their approach to using LLMs and how they’re continuing to fine-tune this work through ongoing feedback from expert editors, targeted prompt engineering, instruction tuning, and scenario-based adversarial testing.
Our latest updates delivered to your inbox
Subscribe to our newsletter to keep up with Meta AI news, events, research breakthroughs, and more.
Join us in the pursuit of what’s possible with AI.
Foundational models
Latest news
Foundational models