Sep 22, 2026
Who Are We Recommending To? Recommender Systems in the Agentic Web
TL;DR: Recommender systems have always assumed a human recipient but AI agents break that assumption. In our...
Sep 25, 2026
Bootstrapping Conversational Recommendation Agents at Spotify: Synthetic Data Generation and Self-Improvement Loops
Imagine asking Spotify to “play some fresh indie soul I might like” and then, mid-session, saying “I love...
Sep 22, 2026
From IR to RecSys: Evaluating LLM-based Judges in Cranfield-style Recommendation Collections
How do you know which recommender system is actually best? The standard approach of holding out some user...
Sep 22, 2026
Hypothesis-Driven Shelf Generation for Personalised Recommendation
TL;DR: We investigate whether personalised shelf concepts can be generated from listener profiles and...
Latest Publications
We publish research papers and present our work in a wide range of venues.
More publicationsRobust Fusion of Semantic and Behavioural Signals for LLM Reranking in Personalised Search
Aleksandr V. Petrov, Nathan Stein, Erik Lybecker, Emma Schüldt, Daniel Lazarovski, Hugues Bouchard, Mounia Lalmas
Intent-Description Anchoring Bias in LLM-as-a-Judge Evaluation of Recommendation Systems
Himan Abdollahpouri, Kyle Kretschman, Mounia Lalmas
Incremental Recommendation via Causal Models
Athanasios Vlontzos, David Gustafsson, Michael O'Riordan, Ciarán Lee
Research Areas
How do we create more personalized experiences? What can we learn about listeners on how they use written language? How do we optimize testing methodologies? Explore all our research areas below.
Algorithmic Responsibility
Artificial Intelligence
Audio and Visual Intelligence
Causal Inference
Economics
Evaluation
Search & Recommendations
Speech and NLP
User Modeling
We are looking for pioneers to join us in all research areas
We're expanding knowledge of audio and video technology every day, sharing open source frameworks, tools, libraries, and models for everything from research exploration to large-scale production deployment.
Join Us
