2026· Annual Meeting of the Association for Computational Linguistics· pp. 38662-38679· 1 citation· 22 references
Computer Science
TL;DR
This work presents Q U DAR, a dual-perspective adaptive retrieval framework motivated by a systematic analysis of retrieval behavior across retriever type and query format, enabling effective integration of complementary retrieval signals while mitigating noise.
Abstract
Retrieval-augmented generation (RAG) systems depend on retrieval modules to supply grounding evidence for large language models. While hybrid approaches combining sparse and dense retrievers improve performance, most rely on fixed weights that ignore query-specific and corpus-specific variation. Similarly, query expansion has long been used to enrich recall, but its integration with original queries is usually static and can introduce noise. We present Q U DAR, a dual-perspective adaptive retrieval framework motivated by a systematic analysis of retrieval behavior across retriever type (sparse vs. dense) and query format (original vs. expanded). Leveraging margin-derived confidence (e.g., top-1–top-2 score gaps) and LLM-based relevance scoring, Q U DAR dynamically assigns query-specific weights, enabling effective integration of complementary retrieval signals while mitigating noise. Q U DAR is lightweight, retriever-agnostic, and broadly applicable. Experiments show consistent gains over static baselines, improving retrieval quality by 12 – 16 % and yielding more stable performance across queries.
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
This work introduces AnchorQE, a training-free method that separately encodes the original query and its expansion before interpolating them, and shows that AnchorQE improves retrieval effectiveness by up to 12.89% when compared to widely-used expansion-only or text-level concatenation baselines across TREC-DL, LoTTE, and BEIR.
Retrieval-Augmented Generation (RAG) systems face a fundamental trade-off: query expansion can improve retrieval effectiveness for ambiguous or underspecified queries, yet indiscriminate expansion introduces unnecessary latency and retrieval noise. Existing RAG pipelines typically apply expansion uniformly, failing to distinguish between easy and retrieval-challenging queries.To address this issue, we propose DARE-RAG, an adaptive retrieval framework that activates LLM-based query expansion only for retrieval-challenging queries. Our method formulates expansion activation as a lightweight binary classification problem using probe retrieval signals, including score margin, variance, entropy, query length, and lexical specificity. A lightweight MLP predicts whether expansion is likely to improve retrieval quality, and expansion is triggered only when the predicted confidence exceeds a percentile-calibrated threshold.DARE-RAG further integrates a dual-path hybrid retrieval architecture combining BM25 sparse retrieval and BGE dense retrieval, fused via Reciprocal Rank Fusion (RRF), followed by a Cross-Encoder reranker for context refinement. Experiments on NQ-Open and HotpotQA demonstrate that DARE-RAG consistently improves retrieval effectiveness and end-to-end QA accuracy while clearly reducing average end-to-end latency compared with corresponding always-expand variants of BM25, BGE-m3, and their RRF-fused hybrid retriever. Extensive ablation studies and efficiency analyses verify the effectiveness of our utility-guided expansion strategy.
Lixiang Zhu· Poster Volume 0007 The 2026...· 0 citations
LLM-based query expansion improves retrieval by generating document-like passages. In hybrid retrieval, however, most evaluations fuse fixed top-$L$ dense and sparse rankings. Because the cutoff controls both which cross-channel contributions enter fusion and how much of each ranking is accessed, gains measured at one $L$ can change or reverse at another. We separate these effects by evaluating retrieval effectiveness under complete-list fusion and recording the policy-specific per-channel replay stopping depths at which its ordered top-$K$ is certified. We then introduce DESA (Dense Expansion and Sparse Anchoring), a channel-asymmetric query expansion method. An LLM generates complementary reference passages; orthogonal residual expansion adds their new semantic directions to the dense query, while score-product anchoring incorporates their lexical cues into sparse retrieval without broadening the original query's lexical support. Across seven BEIR datasets, DESA improves nDCG@10 and Recall@20 over the unexpanded query by 3.82% and 2.38%, while reducing dense and sparse access depths by 36.90% and 36.56%. With equal dataset weighting, 63.31% of queries become shallower in both channels. However, both depths increase with Contriever on Touch\'e-2020. These results support channel-specific integration of generated passages and joint evaluation of retrieval effectiveness and access depth.
Advancements in retrieval models necessitate re-indexing, a computationally expensive process for large-scale production environments. While updating only the query encoder and continuing to use the old index could be a promising middle ground, dense retrieval systems suffer severe performance drops in this setting. We investigate whether Learned Sparse Retrieval (LSR) can mitigate this backward compatibility issue, as its lexical matching may provide a stable term-based anchor to preserve compatibility across model versions. Experiments on BEIR and the streaming settings of LoTTE show that upgrading only the query encoder causes only a small effectiveness drop in LSR when no mitigation applied, whereas dense retrieval fails severely. We explore lightweight query adaptation methods including ranking fusion, representation fusion, and minimal-training adapters to further improve compatibility. These approaches significantly improve backward compatibility on BEIR and effectively reduce performance loss in streaming retrieval. Code: https://github.com/JingfenQiao/LSR-BC.git
Jingfen Qiao, Gabrielle Poerwawinata, Thong Nguyen et al.· Annual International ACM SIG...· 0 citations
: Faced with the surge of massive natural-language content, information retrieval systems must handle increasingly complex queries while filtering noisy information effectively. Although conventional approaches have made notable progress in matching efficiency and general adaptability, they still struggle to precisely model deep semantic associations between query intent and documents in real-world environments. Such limitations can lead to ranking deviations and omission of critical information. Motivated by recent advances in large language models (LLMs) and their capability to capture deep semantics, we propose DPR-FL , a Dual-Path Retrieval method that integrates fusion-based Filtering with structured LLM Feedback . It combines direct retrieval with a generation-guided retrieval process to form cooperative information flows. By fusing candidate results from multiple sources, applying a high-dimensional semantic filtering strategy, and leveraging LLM-based semantic feedback, DPR-FL refines and optimizes the selection of relevant documents, improving both coverage and relevance. Furthermore, the framework supports adaptive weighting of candidate sources and semantic signals, enhancing robustness in heterogeneous retrieval scenarios. Together, these components enable finer-grained information selection and semantic enrichment, substantially improving retrieval performance and result reliability in complex contexts. Experimental evaluations on standard web search benchmarks, including TREC DL’19 and DL’20, as well as low-resource BEIR datasets, demonstrate that DPR-FL achieves measurable gains across key metrics such as NDCG@10 and MAP, showing improved generalization, robustness, and adaptability in zero-shot retrieval settings.
Wei Jiang, Weichao Zhang, Haocheng Sun et al.· Computers, Materials & C...· 0 citations