Search papers, labs, and topics across Lattice.
LRanker is introduced to address the limitations of LLMs in ranking massive candidate pools by incorporating a candidate aggregation encoder using K-means clustering to model global candidate information. A graph-based test-time scaling mechanism partitions candidates into subsets and generates multiple query embeddings, integrated through an ensemble, to enhance robustness and expressiveness. Experiments on RBench demonstrate LRanker's effectiveness, achieving significant gains across different candidate scales, including a 20-30% improvement in the ultra-large scenario with over 6.8M candidates.
LLMs can now rank millions of candidates with significant accuracy gains thanks to a novel K-means clustering and graph-based ensemble approach that overcomes context length limitations.
Large language models (LLMs) have recently shown strong potential for ranking by capturing semantic relevance and adapting across diverse domains, yet existing methods remain constrained by limited context length and high computational costs, restricting their applicability to real-world scenarios where candidate pools often scale to millions. To address this challenge, we propose LRanker, a framework tailored for large-candidate ranking. LRanker incorporates a candidate aggregation encoder that leverages K-means clustering to explicitly model global candidate information, and a graph-based test-time scaling mechanism that partitions candidates into subsets, generates multiple query embeddings, and integrates them through an ensemble procedure. By aggregating diverse embeddings instead of relying on a single representation, this mechanism enhances robustness and expressiveness, leading to more accurate ranking over massive candidate pools. We evaluate LRanker on seven tasks across three scenarios in RBench with different candidate scales. Experimental results show that LRanker achieves over 30% gains in the RBench-Small scenario, improves by 3-9% in MRR in the RBench-Large scenario, and sustains scalability with 20-30% improvements in the RBench-Ultra scenario with more than 6.8M candidates. Ablation studies further verify the effectiveness of its key components. Together, these findings demonstrate the robustness, scalability, and effectiveness of LRanker for massive-candidate ranking.
