Synthetic query generation has become essential for training dense retrievers, yet prior methods generate one query per document, focusing solely on query quality. We are the first to systematically study multi-query synthesis and discover a quality-diversity trade-off: high-quality queries benefit in-domain tasks, while diverse queries benefit out-of-domain (OOD) generalization. Through controlled experiments on 4 benchmark types across Contriever, RetroMAE, and Qwen3-Embedding, we find that diversity benefit strongly correlates with query complexity (r$\geq$0.95, p<0.05), approximated by content words (CW). We formalize this as the Complexity-Diversity Principle (CDP): query complexity determines optimal diversity. Based on CDP, we propose complexity-aware training: multi-query synthesis for high-complexity tasks and CW-weighted training for existing data. Both strategies improve OOD performance on reasoning-intensive benchmarks, with compounded gains when combined.
Current dense retrievers are not robust to out-of-domain and outlier queries, i.e. their effectiveness on these queries is much poorer than what one would expect. In this paper, we consider a specific instance of such queries: queries that contain typos. We show that a small character level perturbation in queries (as caused by typos) highly impacts the effectiveness of dense retrievers. We then demonstrate that the root cause of this resides in the input tokenization strategy employed by BERT. In BERT, tokenization is performed using the BERT's WordPiece tokenizer and we show that a token wit…
Over the last few years, contextualized pre-trained transformer models such as BERT have provided substantial improvements on information retrieval tasks. Recent approaches based on pre-trained transformer models such as BERT, fine-tune dense low-dimensional contextualized representations of queries and documents in embedding space. While these dense retrievers enjoy substantial retrieval effectiveness improvements compared to sparse retrievers, they are computationally intensive, requiring substantial GPU resources, and dense retrievers are known to be more expensive from both time and resour…
Your Dense Retriever is Secretly an Expeditious Reasoner
School of Computer Science and Engineering, Beihang University, China（北京航空航天大学计算机科学与工程学院）；Beijing Institute for General Artificial Intelligence, China（北京通用人工智能研究院）
ExpandR: Teaching Dense Retrievers Beyond Queries with LLM Guidance
Department of Computer Science and Technology, Northeastern University, China（东北大学计算机科学与技术系）；Department of Computer Science and Technology, Institute for AI, Tsinghua University, China（清华大学人工智能研究院计算机科学与技术系）；Beijing National Research Center for Information Science and Technology, China（北京信息科学与技术国家研究中心）
Recommender retrievers aim to rapidly retrieve a fraction of items from the entire item corpus when a user query requests, with the representative two-tower model trained with the log softmax loss. For efficiently training recommender retrievers on modern hardwares, inbatch sampling, where the items in the mini-batch are shared as negatives to estimate the softmax function, has attained growing interest. However, existing inbatch sampling based strategies just correct the sampling bias of inbatch items with item frequency, being unable to distinguish the user queries within the mini-batch and …
Route Me If You Can: A Benchmark for Query Reformulation Selection
Toronto Metropolitan University（多伦多都会大学）；University of California, Berkeley（加州大学伯克利分校）；University of Waterloo（滑铁卢大学）；Mila - Quebec AI Institute（魁北克人工智能研究所）；University of Toronto（多伦多大学）
Understanding the Behaviors of Environment-aware Information Retrieval
Fudan University（复旦大学）；Alibaba DAMO Academy（阿里巴巴达摩院）；Chinese University of Hong Kong（香港中文大学）；Stanford University（斯坦福大学）；Shanghai University of Finance and Economics（上海财经大学）
FAIR at Meta（Meta的FAIR）；University of Washington（华盛顿大学）；National University of Singapore（新加坡国立大学）；Singapore-MIT Alliance for Research（新加坡-麻省理工联合研究联盟）；Allen Institute for Artificial Intelligence（人工智能研究所）；Stanford University（斯坦福大学）；Massachusetts Institute of Technology（麻省理工学院）；University of California, Berkeley（加州大学伯克利分校）
$\texttt{MixGR}$: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity
Technical University of Darmstadt（达姆施塔特工业大学）；Carnegie Mellon University（卡内基梅隆大学）；University of Washington（华盛顿大学）；University of Pennsylvania（宾夕法尼亚大学）；Tencent AI Lab（腾讯AI实验室）
BRIDGE: Multimodal-to-Text Retrieval via Reinforcement-Learned Query Alignment
High institute for computer & information systems（高等计算机与信息系统学院）；Chungbuk National University（忠北大学）；Assiut University（艾斯尤特大学）；University of Innsbruck（因斯布鲁克大学）
Despite the advantages of their low-resource settings, traditional sparse retrievers depend on exact matching approaches between high-dimensional bag-of-words (BoW) representations of both the queries and the collection. As a result, retrieval performance is restricted by semantic discrepancies and vocabulary gaps. On the other hand, transformer-based dense retrievers introduce significant improvements in information retrieval tasks by exploiting low-dimensional contextualized representations of the corpus. While dense retrievers are known for their relative effectiveness, they suffer from low…
Information Retriever (IR) aims to find the relevant documents (e.g. snippets, passages, and articles) to a given query at large scale. IR plays an important role in many tasks such as open domain question answering and dialogue systems, where external knowledge is needed. In the past, searching algorithms based on term matching have been widely used. Recently, neural-based algorithms (termed as neural retrievers) have gained more attention which can mitigate the limitations of traditional methods. Regardless of the success achieved by neural retrievers, they still face many challenges, e.g. s…
Your Retriever Already Knows: Distribution-Shape QPP for RAG Retrieval Sufficiency
Department of Mathematics, FNSPE CTU in Prague（查尔斯理工大学应用科学与工程学院数学系）；Institute of Information Theory and Automation, Czech Academy of Sciences（捷克科学院信息与自动化研究所）
Pseudo Relevance Feedback (PRF) is known to improve the effectiveness of bag-of-words retrievers. At the same time, deep language models have been shown to outperform traditional bag-of-words rerankers. However, it is unclear how to integrate PRF directly with emergent deep language models. In this article, we address this gap by investigating methods for integrating PRF signals into rerankers and dense retrievers based on deep language models. We consider text-based and vector-based PRF approaches, and investigate different ways of combining and scoring relevance signals. An extensive empiric…
A vital step towards the widespread adoption of neural retrieval models is their resource efficiency throughout the training, indexing and query workflows. The neural IR community made great advancements in training effective dual-encoder dense retrieval (DR) models recently. A dense text retrieval model uses a single vector representation per query and passage to score a match, which enables low-latency first stage retrieval with a nearest neighbor search. Increasingly common, training approaches require enormous compute power, as they either conduct negative passage sampling out of a continu…
