Abstract
We investigate algorithmic collective action in transformer-based recommendersystems. Our use case is a music streaming platform where a collective of fansaims to promote the visibility of an underrepresented artist by strategicallyplacing one of their songs in the existing playlists they control. We introducetwo easily implementable strategies to select the position at which to insertthe song with the goal to boost recommendations at test time. The strategiesexploit statistical properties of the learner by targeting discontinuities inthe recommendations, and leveraging the long-tail nature of song distributions.We evaluate the efficacy of our strategies using a publicly availablerecommender system model released by a major music streaming platform. Ourfindings reveal that through strategic placement even small collectives(controlling less than 0.01\% of the training data) can achieve up to$40\times$ more test time recommendations than an average song with the samenumber of training set occurrences. Focusing on the externalities of thestrategy, we find that the recommendations of other songs are largelypreserved, and the newly gained recommendations are distributed across variousartists. Together, our findings demonstrate how carefully designed collectiveaction strategies can be effective while not necessarily being adversarial.