Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval

arXiv:2607.00448v3 Announce Type: replace-cross Abstract: The two-tower model is widely used in the retrieval stage of large-scale recommendation systems, where training typically relies on in-batch and/or out-of-batch negative sampling. These methods, however, tend to produce easy negatives that the model learns quickly and that provide little training signal. This paper proposes a self-supervised, cluster-based hard negative sampling technique that draws negatives from the same semantic cluster as the positive item; in our production deployment the clusters are derived from large language model (LLM) based multimodal content representations, so that intra-cluster items are genuinely similar and yield informative negatives. To make this deployable at industrial scale, we realize the technique in a real-time, end-to-end framework that maintains a live in-memory item pool and draws cluster-based negatives from it on the fly via global out-of-batch sampling (GOOBS). The framework integrates directly into production two-tower training and serving and scales to billions of training examples with minimal computational overhead. Experiments on four public datasets and a 14-day online A/B test in a large-scale production system show that the proposed technique outperforms widely used industry methods, while also helping to break recommendation feedback loops and substantially reducing popularity bias.

Abstract:The two-tower model is widely used in the retrieval stage of large-scale recommendation systems, where training typically relies on in-batch and/or out-of-batch negative sampling. These methods, however, tend to produce easy negatives that the model learns quickly and that provide little training signal. This paper proposes a self-supervised, cluster-based hard negative sampling technique that draws negatives from the same semantic cluster as the positive item; in our production deployment the clusters are derived from large language model (LLM) based multimodal content representations, so that intra-cluster items are genuinely similar and yield informative negatives. To make this deployable at industrial scale, we realize the technique in a real-time, end-to-end framework that maintains a live in-memory item pool and draws cluster-based negatives from it on the fly via global out-of-batch sampling (GOOBS). The framework integrates directly into production two-tower training and serving and scales to billions of training examples with minimal computational overhead. Experiments on four public datasets and a 14-day online A/B test in a large-scale production system show that the proposed technique outperforms widely used industry methods, while also helping to break recommendation feedback loops and substantially reducing popularity bias.

Information Retrieval (cs.IR); Artificial Intelligence (cs.AI)

arXiv:2607.00448 [cs.IR]

(or arXiv:2607.00448v3 [cs.IR] for this version)

https://doi.org/10.48550/arXiv.2607.00448 Focus to learn more arXiv-issued DOI via DataCite

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

aiqbench analysis

Real-Time Hard Negative Sampling via LLM-based Clustering for Large-Scale Two-Tower Retrieval

Related coverage