SERM: Self-Evolving Relevance Model with Agent-Driven Learning from Massive Query Streams
This addresses the challenge of maintaining relevance in search systems for industrial applications with massive, evolving query streams, representing an incremental improvement through multi-agent techniques.
The paper tackles the problem of relevance models struggling with dynamically evolving query streams in large-scale industrial settings by proposing SERM, a Self-Evolving Relevance Model with multi-agent modules for sample mining and annotation, achieving significant performance gains as validated in offline and online tests serving billions of daily requests.
Due to the dynamically evolving nature of real-world query streams, relevance models struggle to generalize to practical search scenarios. A sophisticated solution is self-evolution techniques. However, in large-scale industrial settings with massive query streams, this technique faces two challenges: (1) informative samples are often sparse and difficult to identify, and (2) pseudo-labels generated by the current model could be unreliable. To address these challenges, in this work, we propose a Self-Evolving Relevance Model approach (SERM), which comprises two complementary multi-agent modules: a multi-agent sample miner, designed to detect distributional shifts and identify informative training samples, and a multi-agent relevance annotator, which provides reliable labels through a two-level agreement framework. We evaluate SERM in a large-scale industrial setting, which serves billions of user requests daily. Experimental results demonstrate that SERM can achieve significant performance gains through iterative self-evolution, as validated by extensive offline multilingual evaluations and online testing.