Jul 2026· Annual International ACM SIGIR Conference on Research and Development in Information Retrieval· 0 citations· 28 references
Computer Science
TL;DR
This work introduces HybridSparse, an end-to-end hybrid retrieval framework that strengthens sparse--dense interaction across modeling, training, and serving and adopts a unified encoder with a shared backbone and jointly optimizes lexical and semantic representations through co-training.
Abstract
Large-scale retrieval systems must operate under strict latency constraints while maintaining high recall. Sparse retrieval offers efficiency and interpretability, whereas dense retrieval provides stronger semantic matching. Although hybrid approaches combine both signals, their interaction is often limited, especially under intersection-based retrieval. We introduce HybridSparse, an end-to-end hybrid retrieval framework that strengthens sparse--dense interaction across modeling, training, and serving. It adopts a unified encoder with a shared backbone and jointly optimizes lexical and semantic representations through co-training. To further improve alignment, we incorporate hybrid score regularization and consistency distillation, enabling more stable and effective hybrid scoring. Experiments on public benchmarks demonstrate consistent improvements over strong sparse, dense, and hybrid baselines. In large-scale production deployment for Bing advertisement retrieval, HybridSparse delivers a +1.30% RPM gain, highlighting its practical impact.
A unified pipeline deployed at Walmart that addresses both signal quality and model evolution is presented, and a Warm-Start Distillation technique that transfers domain-specific expertise from the legacy model to the new backbone is introduced.
Zhen Yang, Juexin Lin, Hongwei Shang et al.· Annual International ACM SIG...· 1 citation
This work investigates whether Learned Sparse Retrieval (LSR) can mitigate this backward compatibility issue, and explores lightweight query adaptation methods including ranking fusion, representation fusion, and minimal-training adapters to further improve compatibility.
Jingfen Qiao, Gabrielle Poerwawinata, Thong Nguyen et al.· Annual International ACM SIG...· 0 citations
This work presents DESA (Dense Expansion and Sparse Anchoring), which shares generated references across channels but specializes their integration, which improves nDCG@10 and Recall@20 over the unexpanded query and reduces dense and sparse replay stopping depths.
Generative retrieval (GR) is a promising paradigm for industrial search advertising, yet its deployment is constrained by strict relevance and latency requirements. Existing systems cascade GR with an independent relevance model, decoupling the generative likelihood objective from query-ad relevance discrimination, which compromises effectiveness and increases serving costs. We propose a Unified Generative-Discriminative framework (UniGD) that integrates retrieval and relevance scoring within a single model. To mitigate gradient interference in joint optimization, UniGD introduces Conflict-Aware Gradient Enhancement (CAGE) to adaptively coordinate the two objectives. UniGD further designs a Codebook-Anchored Representation Module (CAM) that anchors item representations to frozen hierarchical codebooks distilled from a multimodal pretrained model, thereby endowing them with rich and generalizable semantic priors. For heterogeneous short-video, product, and live-stream ads, UniGD proposes Heterogeneous Ad-material Modeling (HAM), which captures cross-type semantic commonality over a shared backbone while preserving type-specific modeling capacity. Online AB tests on Kuaishou search advertising platform show that UniGD raises ad revenue by 5.78%, reduces inference latency by 33%, and improves discriminative relevance estimation. On NQ320K and MS300K, UniGD improves Recall@10 over the strongest reproduced GR baseline by 8.44% and 3.19%, respectively.
Shujie Ji, Yawei Kong, Yili Zhao et al.· 0 citations
The Adaptive Multi-Stage Vector Retrieval (AMSVR) framework is proposed, prioritising weighted, drift-resistant composition over uniform fusion, and offers tailored configurations: AMSVR-Scientific (dense + tuned hybrid) peaks at NDCG@10 = 0.7570 on SciFact, while AMSVR-Full (seven stages) targets broader, noisier corpora where Recall@100 matters most.
Samsudeen Alabi Bankole, Yakub Kayode Saheed· NLP & Big Data· 0 citations
Externally-transfer performance after distillation remains mixed, so the evidence supports compression of teacher rankings under matched retrieval protocols.
K. Dubovikov, Martin Takác, S. Lahlou· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
