We consider the large-scale query-document retrieval problem: given a query (e.g., a question), return the set of relevant documents (e.g., paragraphs containing the answer) from a large document corpus. This problem is often solved in two steps. The retrieval phase first reduces the solution space, returning a subset of candidate documents. The scoring phase then re-ranks the documents. Critically, the retrieval algorithm not only desires high recall but also requires to be highly efficient, returning candidates in time sublinear to the number of documents. Unlike the scoring phase witnessing significant advances recently due to the BERT-style pre-training tasks on cross-attention models, the retrieval phase remains less well studied. Most previous works rely on classic Information Retrieval (IR) methods such as BM-25 (token matching + TF-IDF weights). These models only accept sparse handcrafted features and can not be optimized for different downstream tasks of interest. In this paper, we conduct a comprehensive study on the embedding-based retrieval models. We show that the key ingredient of learning a strong embedding-based Transformer model is the set of pre-training tasks. With adequately designed paragraph-level pre-training tasks, the Transformer models can remarkably improve over the widely-used BM-25 as well as embedding models without Transformers. The paragraph-level pre-training tasks we studied are Inverse Cloze Task (ICT), Body First Selection (BFS), Wiki Link Prediction (WLP), and the combination of all three.
In retrieval applications, binary hashes are known to offer significant improvements in terms of both memory and speed. We investigate the compression of sentence embeddings using a neural encoder-decoder architecture, which is trained by minimizing reconstruction error. Instead of employing the original real-valued embeddings, we use latent representations in Hamming space produced by the encoder for similarity calculations. In quantitative experiments on several benchmarks for semantic similarity tasks, we show that our compressed hamming embeddings yield a comparable performance to uncompre…
Research Studio Data Science, RSA FG（RSA FG数据科学研究工作室）
We replicate recent experiments attempting to demonstrate an attractive hypothesis about the use of the Fisher kernel framework and mixture models for aggregating word embeddings towards document representations and the use of these representations in document classification, clustering, and retrieval. Specifically, the hypothesis was that the use of a mixture model of von Mises-Fisher (VMF) distributions instead of Gaussian distributions would be beneficial because of the focus on cosine distances of both VMF and the vector space model traditionally used in information retrieval. Previous exp…
Université de Grenoble（格勒诺布尔大学）；CNRS（法国国家科学研究中心）；Université de Saint-Étienne（圣艾蒂安大学）
This paper presents preliminary works on using Word Embedding (word2vec) for query expansion in the context of Personalized Information Retrieval. Traditionally, word embeddings are learned on a general corpus, like Wikipedia. In this work we try to personalize the word embeddings learning, by achieving the learning on the user's profile. The word embeddings are then in the same context than the user interests. Our proposal is evaluated on the CLEF Social Book Search 2016 collection. The results obtained show that some efforts should be made in the way to apply Word Embedding in the context of…
GESIS - Leibniz Institute for the Social Sciences（莱布尼茨社会科学研究所（GESIS））；Cologne（科隆）
Concepts are used to solve the term-mismatch problem. However, we need an effective similarity measure between concepts. Word embedding presents a promising solution. We present in this study three approaches to build concepts vectors based on words vectors. We use a vector-based measure to estimate inter-concepts similarity. Our experiments show promising results. Furthermore, words and concepts become comparable. This could be used to improve conceptual indexing process.
University of Arkansas（阿肯色大学）；Google LLC.（谷歌有限责任公司）
Deep language models learning a hierarchical representation proved to be a powerful tool for natural language processing, text mining and information retrieval. However, representations that perform well for retrieval must capture semantic meaning at different levels of abstraction or context-scopes. In this paper, we propose a new method to generate multi-resolution word embeddings that represent documents at multiple resolutions in terms of context-scopes. In order to investigate its performance,we use the Stanford Question Answering Dataset (SQuAD) and the Question Answering by Search And R…
Distributed representations of words have shown to be useful to improve the effectiveness of IR systems in many sub-tasks like query expansion, retrieval and ranking. Algorithms like word2vec, GloVe and others are also key factors in many improvements in different NLP tasks. One common issue with such embedding models is that words like happy and sad appear in similar contexts and hence are wrongly clustered close in the embedding space. In this paper we leverage Aff2Vec, a set of word embeddings models which include affect information, in order to better capture the affect aspect in news text…
Lessons Learned on Information Retrieval in Electronic Health Records: A Comparison of Embedding Models and Pooling Strategies
University of Wisconsin-Madison（威斯康星大学麦迪逊分校）；Boston Children’s Hospital Harvard Medical School（波士顿儿童医院哈佛医学院）；University of Colorado Anschutz（科罗拉多大学安舒茨分校）；Loyola University Chicago（芝加哥洛约拉大学）
Bridging the Question-Answer Gap in Retrieval-Augmented Generation: Hypothetical Prompt Embeddings
University of Primorska（普里莫尔斯卡大学）；InnoRenew CoE；Research Centre of the Slovenian Academy of Sciences and Arts（斯洛文尼亚科学与艺术研究院研究中心）；The Fran Ramovš Institute（弗兰·拉莫夫什研究所）
Computer Science and Engineering, Indian Institute of Technology, Kanpur（印度理工学院坎普尔分校计算机科学与工程系）
Cross-lingual information retrieval is a challenging task in the absence of aligned parallel corpora. In this paper, we address this problem by considering topically aligned corpora designed for evaluating an IR setup. To emphasize, we neither use any sentence-aligned corpora or document-aligned corpora, nor do we use any language specific resources such as dictionary, thesaurus, or grammar rules. Instead, we use an embedding into a common space and learn word correspondences directly from there. We test our proposed approach for bilingual IR on standard FIRE datasets for Bangla, Hindi and Eng…
Although more and more language pairs are covered by machine translation services, there are still many pairs that lack translation resources. Cross-language information retrieval (CLIR) is an application which needs translation functionality of a relatively low level of sophistication since current models for information retrieval (IR) are still based on a bag-of-words. The Web provides a vast resource for the automatic construction of parallel corpora which can be used to train statistical translation models automatically. The resulting translation models can be embedded in several ways in a…
Neural network representation learning frameworks have recently shown to be highly effective at a wide range of tasks ranging from radiography interpretation via data-driven diagnostics to clinical decision support. This often superior performance comes at the price of dramatically increased training data requirements that cannot be satisfied in every given institution or scenario. As a means of countering such data sparsity effects, distant supervision alleviates the need for scarce in-domain data by relying on a related, resource-rich, task for training. This study presents an end-to-end neu…
Ping An Technology（平安科技）；Beihang University（北京航空航天大学）
Dual-Encoders is a promising mechanism for answer retrieval in question answering (QA) systems. Currently most conventional Dual-Encoders learn the semantic representations of questions and answers merely through matching score. Researchers proposed to introduce the QA interaction features in scoring function but at the cost of low efficiency in inference stage. To keep independent encoding of questions and answers during inference stage, variational auto-encoder is further introduced to reconstruct answers (questions) from question (answer) embeddings as an auxiliary task to enhance QA intera…
Word embedding or Word2Vec has been successful in offering semantics for text words learned from the context of words. Audio Word2Vec was shown to offer phonetic structures for spoken words (signal segments for words) learned from signals within spoken words. This paper proposes a two-stage framework to perform phonetic-and-semantic embedding on spoken words considering the context of the spoken words. Stage 1 performs phonetic embedding with speaker characteristics disentangled. Stage 2 then performs semantic embedding in addition. We further propose to evaluate the phonetic-and-semantic natu…
Tables contain valuable knowledge in a structured form. We employ neural language modeling approaches to embed tabular data into vector spaces. Specifically, we consider different table elements, such caption, column headings, and cells, for training word and entity embeddings. These embeddings are then utilized in three particular table-related tasks, row population, column population, and table retrieval, by incorporating them into existing retrieval models as additional semantic similarity signals. Evaluation results show that table embeddings can significantly improve upon the performance …
Text embedding representing natural language documents in a semantic vector space can be used for document retrieval using nearest neighbor lookup. In order to study the feasibility of neural models specialized for retrieval in a semantically meaningful way, we suggest the use of the Stanford Question Answering Dataset (SQuAD) in an open-domain question answering context, where the first task is to find paragraphs useful for answering a given question. First, we compare the quality of various text-embedding methods on the performance of retrieval and give an extensive empirical comparison on t…
In this paper, we propose a new approach to learn multimodal multilingual embeddings for matching images and their relevant captions in two languages. We combine two existing objective functions to make images and captions close in a joint embedding space while adapting the alignment of word embeddings between existing languages in our model. We show that our approach enables better generalization, achieving state-of-the-art performance in text-to-image and image-to-text retrieval task, and caption-caption similarity task. Two multimodal multilingual datasets are used for evaluation: Multi30k …
From Topology to Retrieval: Decoding Embedding Spaces with Unified Signatures
University of Tübingen（图宾根大学）；Fribourg University（弗里堡大学）；Boehringer Ingelheim GmbH, Biberach (Riss), Germany（勃林格殷格翰公司）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）
We present an analysis of large-scale pretrained deep learning models used for cross-modal (text-to-audio) retrieval. We use embeddings extracted by these models in a metric learning framework to connect matching pairs of audio and text. Shallow neural networks map the embeddings to a common dimensionality. Our system, which is an extension of our submission to the Language-based Audio Retrieval Task of the DCASE Challenge 2022, employs the RoBERTa foundation model as the text embedding extractor. A pretrained PANNs model extracts the audio embeddings. To improve the generalisation of our mode…
With the wide development of black-box machine learning algorithms, particularly deep neural network (DNN), the practical demand for the reliability assessment is rapidly rising. On the basis of the concept that `Bayesian deep learning knows what it does not know,' the uncertainty of DNN outputs has been investigated as a reliability measure for the classification and regression tasks. However, in the image-caption retrieval task, well-known samples are not always easy-to-retrieve samples. This study investigates two aspects of image-caption embedding-and-retrieval systems. On one hand, we qua…
Query Drift Compensation: Enabling Compatibility in Continual Learning of Retrieval Embedding Models
Computer Vision Center, Barcelona, Spain（巴塞罗那计算机视觉中心）；Department of Computer Science, Universitat Autònoma de Barcelona, Spain（巴塞罗那自治大学计算机科学系）；IDEAS Research Center, Warsaw, Poland（华沙IDEAS研究中心）
Samsung AI Centre Toronto（多伦多三星人工智能中心）；York University（约克大学）；University of Toronto（多伦多大学）
Probabilistic embeddings have proven useful for capturing polysemous word meanings, as well as ambiguity in image matching. In this paper, we study the advantages of probabilistic embeddings in a cross-modal setting (i.e., text and images), and propose a simple approach that replaces the standard vector point embeddings in extant image-text matching models with probabilistic distributions that are parametrically learned. Our guiding hypothesis is that the uncertainty encoded in the probabilistic embeddings captures the cross-modal ambiguity in the input instances, and that it is through captur…
Deep neural networks trained for classification have been found to learn powerful image representations, which are also often used for other tasks such as comparing images w.r.t. their visual similarity. However, visual similarity does not imply semantic similarity. In order to learn semantically discriminative features, we propose to map images onto class embeddings whose pair-wise dot products correspond to a measure of semantic similarity between classes. Such an embedding does not only improve image retrieval results, but could also facilitate integrating semantics for other tasks, e.g., n…
Embedding-based Retrieval (EBR) in e-commerce search is a powerful search retrieval technique to address semantic matches between search queries and products. However, commercial search engines like Facebook Marketplace Search are complex multi-stage systems optimized for multiple business objectives. At Facebook Marketplace, search retrieval focuses on matching search queries with relevant products, while search ranking puts more emphasis on contextual signals to up-rank the more engaging products. As a result, the end-to-end searcher experience is a function of both relevance and engagement,…
University of Toronto（多伦多大学）；Vector Institute（矢量研究院）；Google Brain（谷歌大脑）
We present a new technique for learning visual-semantic embeddings for cross-modal retrieval. Inspired by hard negative mining, the use of hard negatives in structured prediction, and ranking loss functions, we introduce a simple change to common loss functions used for multi-modal embeddings. That, combined with fine-tuning and use of augmented data, yields significant gains in retrieval performance. We showcase our approach, VSE++, on MS-COCO and Flickr30K datasets, using ablation studies and comparisons with existing methods. On MS-COCO our approach outperforms state-of-the-art methods by 8…
Joint modeling of language and vision has been drawing increasing interest. A multimodal data representation allowing for bidirectional retrieval of images by sentences and vice versa is a key aspect. In this paper we present three contributions in canonical correlation analysis (CCA) based multimodal retrieval. Firstly, we show that an asymmetric weighting of the canonical weights, while achieving a cross view mapping from the search to the query space, improves the retrieval performance. Secondly, we devise a computationally efficient model selection, crucial to generalization and stability,…
In this paper, the problem of multi-view embedding from different visual cues and modalities is considered. We propose a unified solution for subspace learning methods using the Rayleigh quotient, which is extensible for multiple views, supervised learning, and non-linear embeddings. Numerous methods including Canonical Correlation Analysis, Partial Least Sqaure regression and Linear Discriminant Analysis are studied using specific intrinsic and penalty graphs within the same framework. Non-linear extensions based on kernels and (deep) neural networks are derived, achieving better performance …
University of Tübingen（图宾根大学）；MPI for Informatics（马克斯·普朗克信息学研究所）；MPI for Intelligent Systems（马克斯·普朗克智能系统研究所）
Existing works in image retrieval often consider retrieving images with one or two query inputs, which do not generalize to multiple queries. In this work, we investigate a more challenging scenario for composing multiple multimodal queries in image retrieval. Given an arbitrary number of query images and (or) texts, our goal is to retrieve target images containing the semantic concepts specified in multiple multimodal queries. To learn an informative embedding that can flexibly encode the semantics of various queries, we propose a novel multimodal probabilistic composer (MPC). Specifically, w…
University of Bristol（布里斯托大学）；NAVER LABS Europe（NAVER LABS 欧洲）
Given a gallery of uncaptioned video sequences, this paper considers the task of retrieving videos based on their relevance to an unseen text query. To compensate for the lack of annotations, we rely instead on a related video gallery composed of video-caption pairs, termed the source gallery, albeit with a domain gap between its videos and those in the target gallery. We thus introduce the problem of Unsupervised Domain Adaptation for Cross-modal Video Retrieval, along with a new benchmark on fine-grained actions. We propose a novel iterative domain alignment method by means of pseudo-labelli…
