Measuring effectiveness and efficiency in information retrieval has a strong empirical background. While modern retrieval systems substantially improve effectiveness, the community has not yet agreed on how to measure efficiency, making it difficult to contrast effectiveness and efficiency fairly. Efficiency-oriented system comparisons are difficult due to factors such as hardware configurations, software versioning, and experimental settings. Efficiency affects users, researchers, and the environment and can be measured in many dimensions beyond time and space, such as resource consumption, water usage, and sample efficiency. Analyzing the efficiency of algorithms and their trade-off with effectiveness requires revisiting and establishing new standards and principles, from defining relevant concepts to designing new measures and guidelines to assess the findings' significance. ReNeuIR's fourth iteration aims to bring the community together to debate these questions and collaboratively test and improve benchmarking frameworks for efficiency based on discussions and collaborations of its previous iterations, including a shared task focused on efficiency and reproducibility.

Whether a text document is freed from the rules of grammar, stripped of word order, and thereby turned into a bag of words, or whether its semantic nuances learnt and condensed into an embedding space, its final representation is the same mathematical object: a vector. In fact, vectors represent much more than just text documents. Any object, be it a document or query, that contains text, images, speech, or a mix of these modalities, is often represented as a vector. Collect a large enough quantity of these vectors and the fundamental question of retrieval from the Information Retrieval (IR) discipline becomes urgently relevant: Finding k vectors that are more similar to a query. This full-day tutorial is concerned with the question above and intends to cover foundational concepts and advanced algorithms for vector retrieval or vector search. The tutorial begins with a focus on foundational concepts, including a brief history from space partitioning, to locality-sensitive hashing, graph-based, and clustering-based methods. As we discuss each class of solutions, we show failure scenarios and explain why they prove insufficient. We conclude the tutorial by turning our attention in the second half to recent developments for maximum inner product search over dense and sparse vectors, as well as open questions that need further research. Through this tutorial, we wish to recap the fascinating topic of retrieval in modern IR for the community, lower barriers of entry into this rich area of research, and inspire interest in conducting research on the underlying theoretical and empirical questions that are specific to IR.

Learned sparse representations form an effective and interpretable class of embeddings for text retrieval. While exact top-k retrieval over such embeddings faces efficiency challenges, a recent algorithm called Seismic has enabled remarkably fast, highly-accurate approximate retrieval. Seismic statically prunes inverted lists, organizes each list into geometrically-cohesive blocks, and augments each block with a summary vector. At query time, each inverted list associated with a query term is traversed one block at a time in an arbitrary order, with the inner product between the query and summaries determining if a block must be evaluated. When a block is deemed promising, its documents are fully evaluated with a forward index. Seismic is one to two orders of magnitude faster than state-of-the-art inverted index-based solutions and significantly outperforms the winning graph-based submissions to the BigANN 2023 Challenge. In this work, we speed up Seismic further by introducing two innovations to its query processing subroutine. First, we traverse blocks in order of importance, rather than arbitrarily. Second, we take the list of documents retrieved by Seismic and expand it to include the neighbors of each document using an offline k-regular nearest neighbor graph; the expanded list is then ranked to produce the final top-k set. Experiments on two public datasets show that our extension, named SeismicWave, can reach almost-exact accuracy levels and is up to 2.2x faster than Seismic.

Learned sparse representations form an attractive class of contextual embeddings for text retrieval. That is so because they are effective models of relevance and are interpretable by design. Despite their apparent compatibility with inverted indexes, however, retrieval over sparse embeddings remains challenging. That is due to the distributional differences between learned embeddings and term frequency-based lexical models of relevance such as BM25. Recognizing this challenge, a great deal of research has gone into, among other things, designing retrieval algorithms tailored to the properties of learned sparse representations, including approximate retrieval systems. In fact, this task featured prominently in the latest BigANN Challenge at NeurIPS 2023, where approximate algorithms were evaluated on a large benchmark dataset by throughput and recall. In this work, we propose a novel organization of the inverted index that enables fast yet effective approximate retrieval over learned sparse embeddings. Our approach organizes inverted lists into geometrically-cohesive blocks, each equipped with a summary vector. During query processing, we quickly determine if a block must be evaluated using the summaries. As we show experimentally, single-threaded query processing using our method, Seismic, reaches sub-millisecond per-query latency on various sparse embeddings of the MS MARCO dataset while maintaining high recall. Our results indicate that Seismic is one to two orders of magnitude faster than state-of-the-art inverted index-based solutions and further outperforms the winning (graph-based) submissions to the BigANN Challenge by a significant margin.

A critical piece of the modern information retrieval puzzle is approximate nearest neighbor search. Its objective is to return a set of k data points that are closest to a query point, with its accuracy measured by the proportion of exact nearest neighbors captured in the returned set. One popular approach to this question is clustering: The indexing algorithm partitions data points into non-overlapping subsets and represents each partition by a point such as its centroid. The query processing algorithm first identifies the nearest clusters---a process known as routing---then performs a nearest neighbor search over those clusters only. In this work, we make a simple observation: The routing function solves a ranking problem. Its quality can therefore be assessed with a ranking metric, making the function amenable to learning-to-rank. Interestingly, ground-truth is often freely available: Given a query distribution in a top-k configuration, the ground-truth is the set of clusters that contain the exact top-k vectors. We develop this insight and apply it to Maximum Inner Product Search (MIPS). As we demonstrate empirically on various datasets, learning a simple linear function consistently improves the accuracy of clustering-based MIPS.

We study hybrid search in text retrieval where lexical and semantic search are fused together with the intuition that the two are complementary in how they model relevance. In particular, we examine fusion by a convex combination of lexical and semantic scores, as well as the reciprocal rank fusion (RRF) method, and identify their advantages and potential pitfalls. Contrary to existing studies, we find RRF to be sensitive to its parameters; that the learning of a convex combination fusion is generally agnostic to the choice of score normalization; that convex combination outperforms RRF in in-domain and out-of-domain settings; and finally, that convex combination is sample efficient, requiring only a small set of training examples to tune its only parameter to a target domain.

The aim of this Special Section is to engage with researchers in Information Retrieval, Natural Language Processing and related areas and gather insight into the core challenges in measuring, reporting, and optimizing all facets of efficiency in Neural Information Retrieval (NIR) systems, including time-, space-, resource-, sample-, and energy-efficiency, among other factors. While researchers in the field have assiduously explored the Pareto frontier in quality and efficiency in other contexts for decades, we believe that the neural dimension introduces new hurdles.

Maximum Inner Product Search or top- k retrieval on sparse vectors is well understood in information retrieval, with a number of mature algorithms that solve it exactly. However, all existing algorithms are tailored to text and frequency-based similarity measures. To achieve optimal memory footprint and query latency, they rely on the near stationarity of documents and on laws governing natural languages. We consider, instead, a setup in which collections are streaming—necessitating dynamic indexing—and where indexing and retrieval must work with arbitrarily distributed real-valued vectors. As we show, existing algorithms are no longer competitive in this setup, even against naïve solutions. We investigate this gap and present a novel approximate solution, called Sinnamon , that can efficiently retrieve the top- k results for sparse real valued vectors drawn from arbitrary distributions. Notably, Sinnamon offers levers to trade off memory consumption, latency, and accuracy, making the algorithm suitable for constrained applications and systems. We give theoretical results on the error introduced by the approximate nature of the algorithm and present an empirical evaluation of its performance on two hardware platforms and synthetic and real-valued datasets. We conclude by laying out concrete directions for future research on this general top- k retrieval problem over sparse vectors.

Maximum inner product search (MIPS) over dense and sparse vectors have progressed independently in a bifurcated literature for decades; the latter is better known as top- \(k\) retrieval in Information Retrieval. This duality exists because sparse and dense vectors serve different end goals. That is despite the fact that they are manifestations of the same mathematical problem. In this work, we ask if algorithms for dense vectors could be applied effectively to sparse vectors, particularly those that violate the assumptions underlying top- \(k\) retrieval methods. We study clustering-based approximate MIPS where vectors are partitioned into clusters and only a fraction of clusters are searched during retrieval. We conduct a comprehensive analysis of dimensionality reduction for sparse vectors, and examine standard and spherical k -means for partitioning. Our experiments demonstrate that clustering-based retrieval serves as an efficient solution for sparse MIPS. As byproducts, we identify two research opportunities and explore their potential. First, we cast the clustering-based paradigm as dynamic pruning and turn that insight into a novel organization of the inverted index for approximate MIPS over general sparse vectors. Second, we offer a unified regime for MIPS over vectors that have dense and sparse subspaces, that is robust to query distributions.

Yggdrasil Decision Forests is a library for the training, serving and interpretation of decision forest models, targeted both at research and production work, implemented in C++, and available in C++, command line interface, Python (under the name TensorFlow Decision Forests), JavaScript, Go, and Google Sheets (under the name Simple ML for Sheets). The library has been developed organically since 2018 following a set of four design principles applicable to machine learning libraries and frameworks: simplicity of use, safety of use, modularity and high-level abstraction, and integration with other machine learning libraries. In this paper, we describe those principles in detail and present how they have been used to guide the design of the library. We then showcase the use of our library on a set of classical machine learning problems. Finally, we report a benchmark comparing our library to related solutions.

Multifaceted, empirical evaluation of algorithmic ideas is one of the central pillars of Information Retrieval (IR) research. The IR community has a rich history of studying the effectiveness of indexes, retrieval algorithms, and complex machine learning rankers and, at the same time, quantifying their computational costs, from creation and training to application and inference. As the community moves towards even more complex deep learning models, questions on efficiency have once again become relevant with renewed urgency. Indeed, efficiency is no longer limited to time and space; instead it has found new, challenging dimensions that stretch to resource-, sample- and energy-efficiency with ramifications for researchers, users, and the environment alike. Examining algorithms and models through the lens of holistic efficiency requires the establishment of standards and principles, from defining relevant concepts, to designing metrics, to creating guidelines for making sense of the significance of new findings. The second iteration of the ReNeuIR workshop aims to bring the community together to debate these questions, with the express purpose of moving towards a common benchmarking framework for efficiency.

Perhaps the applied nature of information retrieval research goes some way to explain the community's rich history of evaluating machine learning models holistically, understanding that efficacy matters but so does the computational cost incurred to achieve it. This is evidenced, for example, by more than a decade of research on efficient training and inference of large decision forest models in learning-to-rank. As the community adopts even more complex, neural network-based models in a wide range of applications, questions on efficiency have once again become relevant. We propose this workshop as a forum for a critical discussion of efficiency in the era of neural information retrieval, to encourage debate on the current state and future directions of research in this space, and to promote more sustainable research by identifying best practices in the development and evaluation of neural models for information retrieval.

Listwise learning-to-rank methods form a powerful class of ranking algorithms that are widely adopted in applications such as information retrieval. These algorithms learn to rank a set of items by optimizing a loss that is a function of the entire set—as a surrogate to a typically non-differentiable ranking metric. Despite their empirical success, existing listwise methods are based on heuristics and remain theoretically ill-understood. In particular, none of the empirically successful loss functions are related to ranking metrics. In this work, we propose a cross entropy-based learning-to-rank loss function that is theoretically sound, is a convex bound on NDCG—a popular ranking metric—and is consistent with NDCG under learning scenarios common in information retrieval. Furthermore, empirical evaluation of an implementation of the proposed method with gradient boosting machines on benchmark learning-to-rank datasets demonstrates the superiority of our proposed formulation over existing algorithms in quality and robustness.

Learning to Rank, a central problem in information retrieval, is a class of machine learning algorithms that formulate ranking as an optimization task. The objective is to learn a function that produces an ordering of a set of documents in such a way that the utility of the entire ordered list is maximized. Learning-to-rank methods do so by learning a function that computes a score for each document in the set. A ranked list is then compiled by sorting documents according to their scores. While such a deterministic mapping of scores to permutations makes sense during inference where stability of ranked lists is required, we argue that its greedy nature during training leads to less robust models. This is particularly problematic when the loss function under optimization---in agreement with ranking metrics---largely penalizes incorrect rankings and does not take into account the distribution of raw scores. In this work, we present a stochastic framework where, instead of a deterministic derivation of permutations from raw scores, permutations are sampled from a distribution defined by raw scores. Our proposed sampling method is differentiable and works well with gradient descent optimizers. We analytically study our proposed method and demonstrate when and why it leads to model robustness. We also show empirically, through experiments on publicly available learning-to-rank datasets, that the application of our proposed method to a class of ranking loss functions leads to significant model quality improvements.

Learning-to-Rank deals with maximizing the utility of a list of examples presented to the user, with items of higher relevance being prioritized. It has several practical applications such as large-scale search, recommender systems, document summarization and question answering. While there is widespread support for classification and regression based learning, support for learning-to-rank in deep learning has been limited. We introduce TensorFlow Ranking, the first open source library for solving large-scale ranking problems in a deep learning framework. It is highly configurable and provides easy-to-use APIs to support different scoring mechanisms, loss functions and evaluation metrics in the learning-to-rank setting. Our library is developed on top of TensorFlow and can thus fully leverage the advantages of this platform. TensorFlow Ranking has been deployed in production systems within Google; it is highly scalable, both in training and in inference, and can be used to learn ranking models over massive amounts of user activity data, which can include heterogeneous dense and sparse features. We empirically demonstrate the effectiveness of our library in learning ranking functions for large-scale search and recommendation applications in Gmail and Google Drive. We also show that ranking models built using our model scale well for distributed training, without significant impact on metrics. The proposed library is available to the open source community, with the hope that it facilitates further academic research and industrial applications in the field of learning-to-rank.

Learning-to-Rank is a branch of supervised machine learning that seeks to produce an ordering of a list of items such that the utility of the ranked list is maximized. Unlike most machine learning techniques, however, the objective cannot be directly optimized using gradient descent methods as it is either discontinuous or flat everywhere. As such, learning-to-rank methods often optimize a loss function that either is loosely related to or upper-bounds a ranking utility instead. A notable exception is the approximation framework originally proposed by Qin et al. that facilitates a more direct approach to ranking metric optimization. We revisit that framework almost a decade later in light of recent advances in neural networks and demonstrate its superiority empirically. Through this study, we hope to show that the ideas from that work are more relevant than ever and can lay the foundation of learning-to-rank research in the age of deep neural networks.

This tutorial aims to weave together diverse strands of modern Learning to Rank (LtR) research, and present them in a unified full-day tutorial. First, we will introduce the fundamentals of LtR, and an overview of its various sub-fields. Then, we will discuss some recent advances in gradient boosting methods such as LambdaMART by focusing on their efficiency/effectiveness trade-offs and optimizations. Subsequently, we will then present TF-Ranking, a new open source TensorFlow package for neural LtR models, and how it can be used for modeling sparse textual features. Finally, we will conclude the tutorial by covering unbiased LtR -- a new research field aiming at learning from biased implicit user feedback. The tutorial will consist of three two-hour sessions, each focusing on one of the topics described above. It will provide a mix of theoretical and hands-on sessions, and should benefit both academics interested in learning more about the current state-of-the-art in LtR, as well as practitioners who want to use LtR techniques in their applications.

Tree-based models have proven to be an effective solution for web ranking as well as other machine learning problems in diverse domains. This paper focuses on optimizing the runtime performance of applying such models to make predictions, specifically using gradient-boosted regression trees for learning to rank. Although exceedingly simple conceptually, most implementations of tree-based models do not efficiently utilize modern superscalar processors. By laying out data structures in memory in a more cache-conscious fashion, removing branches from the execution flow using a technique called predication, and micro-batching predictions using a technique called vectorization, we are able to better exploit modern processor architectures. Experiments on synthetic data and on three standard learning-to-rank datasets show that our approach is significantly faster than standard implementations.

We explore a real-time Twitter search application where tweets are arriving at a rate of several thousands per second. Real-time search demands that they be indexed and searchable immediately, which leads to a number of implementation challenges. In this paper, we focus on one aspect: dynamic postings allocation policies for index structures that are completely held in main memory. The core issue can be characterized as a "Goldilocks Problem". Because memory remains today a scare resource, an allocation policy that is too aggressive leads to inefficient utilization, while a policy that is too conservative is slow and leads to fragmented postings lists. We present a dynamic postings allocation policy that allocates memory in increasingly-larger "slices" from a small number of large, fixed pools of memory. With an analytical model and experiments, we explore different settings that balance time (query evaluation speed) and space (memory utilization).

This paper examines a multi-stage retrieval architecture consisting of a candidate generation stage, a feature extraction stage, and a reranking stage using machine-learned models. Given a fixed set of features and a learning-to-rank model, we explore effectiveness/efficiency tradeoffs with three candidate generation approaches: postings intersection with SvS, conjunctive query evaluation with WAND, and disjunctive query evaluation with WAND. We find no significant differences in end-to-end effectiveness as measured by NDCG between conjunctive and disjunctive WAND, but conjunctive query evaluation is substantially faster. Postings intersection with SvS, while fast, yields substantially lower end-to-end effectiveness, suggesting that document and term frequencies remain important in the initial ranking stage. These findings show that conjunctive WAND is the best overall candidate generation strategy of those we examined.

The rise of social media and other forms of user-generated content have created the demand for real-time search: against a high-velocity stream of incoming documents, users desire a list of relevant results at the time the query is issued. In the context of real-time search on tweets, this work explores candidate generation in a two-stage retrieval architecture where an initial list of results is processed by a second-stage rescorer to produce the final output. We introduce Bloom filter chains, a novel extension of Bloom filters that can dynamically expand to efficiently represent an arbitrarily long and growing list of monotonically-increasing integers with a constant false positive rate. Using a collection of Bloom filter chains, a novel approximate candidate generation algorithm called BWand is able to perform both conjunctive and disjunctive retrieval. Experiments show that our algorithm is many times faster than competitive baselines and that this increased performance does not require sacrificing end-to-end effectiveness. Our results empirically characterize the trade-off space defined by output quality, query evaluation speed, and memory footprint for this particular search architecture.

Most modern web search engines employ a two-phase ranking strategy: a candidate list of documents is generated using a "cheap" but low-quality scoring function, which is then reranked by an "expensive" but high-quality method (usually machine-learned). This paper focuses on the problem of candidate generation for conjunctive query processing in this context. We describe and evaluate a fast, approximate postings list intersection algorithms based on Bloom filters. Due to the power of modern learning-to-rank techniques and emphasis on early precision, significant speedups can be achieved without loss of end-to-end retrieval effectiveness. Explorations reveal a rich design space where effectiveness and efficiency can be balanced in response to specific hardware configurations and application scenarios.

Latent Dirichlet Allocation (LDA) is a popular topic modeling technique for exploring document collections. Because of the increasing prevalence of large datasets, there is a need to improve the scalability of inference for LDA. In this paper, we introduce a novel and flexible large scale topic modeling package in MapReduce (Mr. LDA). As opposed to other techniques which use Gibbs sampling, our proposed framework uses variational inference, which easily fits into a distributed environment. More importantly, this variational implementation, unlike highly tuned and specialized implementations based on Gibbs sampling, is easily extensible. We demonstrate two extensions of the models possible with this scalable framework: informed priors to guide topic discovery and extracting topics from a multilingual corpus. We compare the scalability of Mr. LDA against Mahout, an existing large scale topic modeling package. Mr. LDA out-performs Mahout both in execution speed and held-out likelihood.

Test collections are the primary drivers of progress in information retrieval. They provide yardsticks for assessing the effectiveness of ranking functions in an automatic, rapid, and repeatable fashion and serve as training data for learning to rank models. However, manual construction of test collections tends to be slow, labor-intensive, and expensive. This paper examines the feasibility of constructing web search test collections in a completely unsupervised manner given only a large web corpus as input. Within our proposed framework, anchor text extracted from the web graph is treated as a pseudo query log from which pseudo queries are sampled. For each pseudo query, a set of relevant and non-relevant documents are selected using a variety of web-specific features, including spam and aggregated anchor text weights. The automatically mined queries and judgments form a pseudo test collection that can be used for training ranking functions. Experiments carried out on TREC web track data show that learning to rank models trained using pseudo test collections outperform an unsupervised ranking function and are statistically indistinguishable from a model trained using manual judgments, demonstrating the usefulness of our approach in extracting reasonable quality training data "for free".

No abstract available.