We introduce an approach for open-domain question answering (QA) that retrieves and reads a passage graph, where vertices are passages of text and edges represent relationships that are derived from an external knowledge base or co-occurrence in the same article. Our goals are to boost coverage by using knowledge-guided retrieval to find more relevant passages than text-matching methods, and to improve accuracy by allowing for better knowledge-guided fusion of information across related passages. Our graph retrieval method expands a set of seed keyword-retrieved passages by traversing the graph structure of the knowledge base. Our reader extends a BERT-based architecture and updates passage representations by propagating information from related passages and their relations, instead of reading each passage in isolation. Experiments on three open-domain QA datasets, WebQuestions, Natural Questions and TriviaQA, show improved performance over non-graph baselines by 2-11% absolute. Our approach also matches or exceeds the state-of-the-art in every case, without using an expensive end-to-end training regime.
Image-text retrieval of natural scenes has been a popular research topic. Since image and text are heterogeneous cross-modal data, one of the key challenges is how to learn comprehensive yet unified representations to express the multi-modal data. A natural scene image mainly involves two kinds of visual concepts, objects and their relationships, which are equally essential to image-text retrieval. Therefore, a good representation should account for both of them. In the light of recent success of scene graph in many CV and NLP tasks for describing complex natural scenes, we propose to represent image and text with two kinds of scene graphs: visual scene graph (VSG) and textual scene graph (TSG), each of which is exploited to jointly characterize objects and relationships in the corresponding modality. The image-text retrieval task is then naturally formulated as cross-modal scene graph matching. Specifically, we design two particular scene graph encoders in our model for VSG and TSG, which can refine the representation of each node on the graph by aggregating neighborhood information. As a result, both object-level and relationship-level cross-modal features can be obtained, which favorably enables us to evaluate the similarity of image and text in the two levels in a more plausible way. We achieve state-of-the-art results on Flickr30k and MSCOCO, which verifies the advantages of our graph matching based approach for image-text retrieval.
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 with German and English captions and Microsoft-COCO with English and Japanese captions.
Joint image-text embedding is the bedrock for most Vision-and-Language (V+L) tasks, where multimodality inputs are simultaneously processed for joint visual and textual understanding. In this paper, we introduce UNITER, a UNiversal Image-TExt Representation, learned through large-scale pre-training over four image-text datasets (COCO, Visual Genome, Conceptual Captions, and SBU Captions), which can power heterogeneous downstream V+L tasks with joint multimodal embeddings. We design four pre-training tasks: Masked Language Modeling (MLM), Masked Region Modeling (MRM, with three variants), Image-Text Matching (ITM), and Word-Region Alignment (WRA). Different from previous work that applies joint random masking to both modalities, we use conditional masking on pre-training tasks (i.e., masked language/region modeling is conditioned on full observation of image/text). In addition to ITM for global image-text alignment, we also propose WRA via the use of Optimal Transport (OT) to explicitly encourage fine-grained alignment between words and image regions during pre-training. Comprehensive analysis shows that both conditional masking and OT-based WRA contribute to better pre-training. We also conduct a thorough ablation study to find an optimal combination of pre-training tasks. Extensive experiments show that UNITER achieves new state of the art across six V+L tasks (over nine datasets), including Visual Question Answering, Image-Text Retrieval, Referring Expression Comprehension, Visual Commonsense Reasoning, Visual Entailment, and NLVR$^2$. Code is available at https://github.com/ChenRocks/UNITER.
Text retrieval systems often return large sets of documents, particularly when applied to large collections. Stopping criteria can reduce the number of these documents that need to be manually evaluated for relevance by predicting when a suitable level of recall has been achieved. In this work, a novel method for determining a stopping criterion is proposed that models the rate at which relevant documents occur using a Poisson process. This method allows a user to specify both a minimum desired level of recall to achieve and a desired probability of having achieved it. We evaluate our method on a public dataset and compare it with previous techniques for determining stopping criteria.
Latent factor models (LFMs) such as matrix factorization have achieved the state-of-the-art performance among various collaborative filtering approaches for recommendation. Despite the high recommendation accuracy of LFMs, a critical issue to be resolved is their lack of interpretability. Extensive efforts have been devoted to interpreting the prediction results of LFMs. However, they either rely on auxiliary information which may not be available in practice, or sacrifice recommendation accuracy for interpretability. Influence functions, stemming from robust statistics, have been developed to understand the effect of training points on the predictions of black-box models. Inspired by this, we propose a novel explanation method named FIA (Fast Influence Analysis) to understand the prediction of trained LFMs by tracing back to the training data with influence functions. We present how to employ influence functions to measure the impact of historical user-item interactions on the prediction results of LFMs and provide intuitive neighbor-style explanations based on the most influential interactions. Our proposed FIA exploits the characteristics of two important LFMs, matrix factorization and neural collaborative filtering, and is capable of accelerating the overall influence analysis process. We provide a detailed complexity analysis for FIA over LFMs and conduct extensive experiments to evaluate its performance using real-world datasets. The results demonstrate the effectiveness and efficiency of FIA, and the usefulness of the generated explanations for the recommendation results.
We propose Unicoder-VL, a universal encoder that aims to learn joint representations of vision and language in a pre-training manner. Borrow ideas from cross-lingual pre-trained models, such as XLM and Unicoder, both visual and linguistic contents are fed into a multi-layer Transformer for the cross-modal pre-training, where three pre-trained tasks are employed, including Masked Language Modeling (MLM), Masked Object Classification (MOC) and Visual-linguistic Matching (VLM). The first two tasks learn context-aware representations for input tokens based on linguistic and visual contents jointly. The last task tries to predict whether an image and a text describe each other. After pretraining on large-scale image-caption pairs, we transfer Unicoder-VL to caption-based image-text retrieval and visual commonsense reasoning, with just one additional output layer. We achieve state-of-the-art or comparable results on both two tasks and show the powerful ability of the cross-modal pre-training.
Transforming XML documents with conventional XML languages, like XSL-T, is disadvantageous because there is too lax abstraction on the target language and it is rather difficult to recognize rule-oriented transformations. Prolog as a programming language of declarative paradigm is especially good for implementation of analysis of formal languages. Prolog seems also to be good for term manipulation, complex schema-transformation and text retrieval. In this report an appropriate model for XML documents is proposed, the basic transformation language for Prolog LTL is defined and the expressiveness power compared with XSL-T is demonstrated, the implementations used throughout are multi paradigmatic.
Visual-semantic embedding aims to find a shared latent space where related visual and textual instances are close to each other. Most current methods learn injective embedding functions that map an instance to a single point in the shared space. Unfortunately, injective embedding cannot effectively handle polysemous instances with multiple possible meanings; at best, it would find an average representation of different meanings. This hinders its use in real-world scenarios where individual instances and their cross-modal associations are often ambiguous. In this work, we introduce Polysemous Instance Embedding Networks (PIE-Nets) that compute multiple and diverse representations of an instance by combining global context with locally-guided features via multi-head self-attention and residual learning. To learn visual-semantic embedding, we tie-up two PIE-Nets and optimize them jointly in the multiple instance learning framework. Most existing work on cross-modal retrieval focuses on image-text data. Here, we also tackle a more challenging case of video-text retrieval. To facilitate further research in video-text retrieval, we release a new dataset of 50K video-sentence pairs collected from social media, dubbed MRW (my reaction when). We demonstrate our approach on both image-text and video-text retrieval scenarios using MS-COCO, TGIF, and our new MRW dataset.
We propose the Neuro-Symbolic Concept Learner (NS-CL), a model that learns visual concepts, words, and semantic parsing of sentences without explicit supervision on any of them; instead, our model learns by simply looking at images and reading paired questions and answers. Our model builds an object-based scene representation and translates sentences into executable, symbolic programs. To bridge the learning of two modules, we use a neuro-symbolic reasoning module that executes these programs on the latent scene representation. Analogical to human concept learning, the perception module learns visual concepts based on the language description of the object being referred to. Meanwhile, the learned visual concepts facilitate learning new words and parsing new sentences. We use curriculum learning to guide the searching over the large compositional space of images and language. Extensive experiments demonstrate the accuracy and efficiency of our model on learning visual concepts, word representations, and semantic parsing of sentences. Further, our method allows easy generalization to new object attributes, compositions, language concepts, scenes and questions, and even new program domains. It also empowers applications including visual question answering and bidirectional image-text retrieval.
Intelligent personal assistant systems that are able to have multi-turn conversations with human users are becoming increasingly popular. Most previous research has been focused on using either retrieval-based or generation-based methods to develop such systems. Retrieval-based methods have the advantage of returning fluent and informative responses with great diversity. However, the performance of the methods is limited by the size of the response repository. On the other hand, generation-based methods can produce highly coherent responses on any topics. But the generated responses are often generic and not informative due to the lack of grounding knowledge. In this paper, we propose a hybrid neural conversation model that combines the merits of both response retrieval and generation methods. Experimental results on Twitter and Foursquare data show that the proposed model outperforms both retrieval-based methods and generation-based methods (including a recently proposed knowledge-grounded neural conversation model) under both automatic evaluation metrics and human evaluation. We hope that the findings in this study provide new insights on how to integrate text retrieval and text generation models for building conversation systems.
The Precision Medicine Initiative states that treatments for a patient should take into account not only the patient's disease, but his/her specific genetic variation as well. The vast biomedical literature holds the potential for physicians to identify effective treatment options for a cancer patient. However, the complexity and ambiguity of medical terms can result in vocabulary mismatch between the physician's query and the literature. The physician's search intent (finding treatments instead of other types of studies) is difficult to explicitly formulate in a query. Therefore, simple ad hot retrieval approach will suffer from low recall and precision. In this paper, we propose a new retrieval system that helps physicians identify effective treatments in precision medicine. Given a cancer patient with a specific disease, genetic variation, and demographic information, the system aims to identify biomedical publications that report effective treatments. We approach this goal from two directions. First, we expand the original disease and gene terms using biomedical knowledge bases to improve recall of the initial retrieval. We then improve precision by promoting treatment-related publications to the top using a machine learning reranker trained on 2017 Text Retrieval Conference Precision Medicine (PM) track corpus. Batch evaluation results on 2018 PM track corpus show that the proposed approach effectively improves both recall and precision, achieving performance comparable to the top entries on the leaderboard of 2018 PM track.
Hashing has been widely applied to multimodal retrieval on large-scale multimedia data due to its efficiency in computation and storage. In this article, we propose a novel deep semantic multimodal hashing network (DSMHN) for scalable image-text and video-text retrieval. The proposed deep hashing framework leverages 2-D convolutional neural networks (CNN) as the backbone network to capture the spatial information for image-text retrieval, while the 3-D CNN as the backbone network to capture the spatial and temporal information for video-text retrieval. In the DSMHN, two sets of modality-specific hash functions are jointly learned by explicitly preserving both intermodality similarities and intramodality semantic labels. Specifically, with the assumption that the learned hash codes should be optimal for the classification task, two stream networks are jointly trained to learn the hash functions by embedding the semantic labels on the resultant hash codes. Moreover, a unified deep multimodal hashing framework is proposed to learn compact and high-quality hash codes by exploiting the feature representation learning, intermodality similarity-preserving learning, semantic label-preserving learning, and hash function learning with different types of loss functions simultaneously. The proposed DSMHN method is a generic and scalable deep hashing framework for both image-text and video-text retrievals, which can be flexibly integrated with different types of loss functions. We conduct extensive experiments for both single modal- and cross-modal-retrieval tasks on four widely used multimodal-retrieval data sets. Experimental results on both image-text- and video-text-retrieval tasks demonstrate that the DSMHN significantly outperforms the state-of-the-art methods.
Most neural Information Retrieval (Neu-IR) models derive query-to-document ranking scores based on term-level matching. Inspired by TileBars, a classical term distribution visualization method, in this paper, we propose a novel Neu-IR model that handles query-to-document matching at the subtopic and higher levels. Our system first splits the documents into topical segments, "visualizes" the matchings between the query and the segments, and then feeds an interaction matrix into a Neu-IR model, DeepTileBars, to obtain the final ranking scores. DeepTileBars models the relevance signals occurring at different granularities in a document's topic hierarchy. It better captures the discourse structure of a document and thus the matching patterns. Although its design and implementation are light-weight, DeepTileBars outperforms other state-of-the-art Neu-IR models on benchmark datasets including the Text REtrieval Conference (TREC) 2010-2012 Web Tracks and LETOR 4.0.
To achieve the long-term goal of machines being able to engage humans in conversation, our models should captivate the interest of their speaking partners. Communication grounded in images, whereby a dialogue is conducted based on a given photo, is a setup naturally appealing to humans (Hu et al., 2014). In this work we study large-scale architectures and datasets for this goal. We test a set of neural architectures using state-of-the-art image and text representations, considering various ways to fuse the components. To test such models, we collect a dataset of grounded human-human conversations, where speakers are asked to play roles given a provided emotional mood or style, as the use of such traits is also a key factor in engagingness (Guo et al., 2019). Our dataset, Image-Chat, consists of 202k dialogues over 202k images using 215 possible style traits. Automatic metrics and human evaluations of engagingness show the efficacy of our approach; in particular, we obtain state-of-the-art performance on the existing IGC task, and our best performing model is almost on par with humans on the Image-Chat test set (preferred 47.7% of the time).
Textual information found in scene images provides high level semantic information about the image and its context and it can be leveraged for better scene understanding. In this paper we address the problem of scene text retrieval: given a text query, the system must return all images containing the queried text. The novelty of the proposed model consists in the usage of a single shot CNN architecture that predicts at the same time bounding boxes and a compact text representation of the words in them. In this way, the text based image retrieval task can be casted as a simple nearest neighbor search of the query text representation over the outputs of the CNN over the entire image database. Our experiments demonstrate that the proposed architecture outperforms previous state-of-the-art while it offers a significant increase in processing speed.
Cross-modal retrieval between visual data and natural language description remains a long-standing challenge in multimedia. While recent image-text retrieval methods offer great promise by learning deep representations aligned across modalities, most of these methods are plagued by the issue of training with small-scale datasets covering a limited number of images with ground-truth sentences. Moreover, it is extremely expensive to create a larger dataset by annotating millions of images with sentences and may lead to a biased model. Inspired by the recent success of webly supervised learning in deep neural networks, we capitalize on readily-available web images with noisy annotations to learn robust image-text joint representation. Specifically, our main idea is to leverage web images and corresponding tags, along with fully annotated datasets, in training for learning the visual-semantic joint embedding. We propose a two-stage approach for the task that can augment a typical supervised pair-wise ranking loss based formulation with weakly-annotated web images to learn a more robust visual-semantic embedding. Experiments on two standard benchmark datasets demonstrate that our method achieves a significant performance gain in image-text retrieval compared to state-of-the-art approaches.
The problem of similarity search is one of the main problems in computer science. This problem has many applications in text-retrieval, web search, computational biology, bioinformatics and others. Similarity between two data objects can be depicted using a similarity measure or a distance metric. There are numerous distance metrics in the literature, some are used for a particular data type, and others are more general. In this paper we present a new distance metric for sequential data which is based on the sum of n-grams. The novelty of our distance is that these n-grams are weighted using artificial bee colony; a recent optimization algorithm based on the collective intelligence of a swarm of bees on their search for nectar. This algorithm has been used in optimizing a large number of numerical problems. We validate the new distance experimentally.
The computation of relatedness between two fragments of text in an automated manner requires taking into account a wide range of factors pertaining to the meaning the two fragments convey, and the pairwise relations between their words. Without doubt, a measure of relatedness between text segments must take into account both the lexical and the semantic relatedness between words. Such a measure that captures well both aspects of text relatedness may help in many tasks, such as text retrieval, classification and clustering. In this paper we present a new approach for measuring the semantic relatedness between words based on their implicit semantic links. The approach exploits only a word thesaurus in order to devise implicit semantic links between words. Based on this approach, we introduce Omiotis, a new measure of semantic relatedness between texts which capitalizes on the word-to-word semantic relatedness measure (SR) and extends it to measure the relatedness between texts. We gradually validate our method: we first evaluate the performance of the semantic relatedness measure between individual words, covering word-to-word similarity and relatedness, synonym identification and word analogy; then, we proceed with evaluating the performance of our method in measuring text-to-text semantic relatedness in two tasks, namely sentence-to-sentence similarity and paraphrase recognition. Experimental evaluation shows that the proposed method outperforms every lexicon-based method of semantic relatedness in the selected tasks and the used data sets, and competes well against corpus-based and hybrid approaches.
A new class of applications based on visual search engines are emerging, especially on smart-phones that have evolved into powerful tools for processing images and videos. The state-of-the-art algorithms for large visual content recognition and content based similarity search today use the "Bag of Features" (BoF) or "Bag of Words" (BoW) approach. The idea, borrowed from text retrieval, enables the use of inverted files. A very well known issue with this approach is that the query images, as well as the stored data, are described with thousands of words. This poses obvious efficiency problems when using inverted files to perform efficient image matching. In this paper, we propose and compare various techniques to reduce the number of words describing an image to improve efficiency and we study the effects of this reduction on effectiveness in landmark recognition and retrieval scenarios. We show that very relevant improvement in performance are achievable still preserving the advantages of the BoF base approach.