As modern data pipelines continue to collect, produce, and store a variety of data formats, extracting and combining value from traditional and context-rich sources such as strings, text, video, audio, and logs becomes a manual process where such formats are unsuitable for RDBMS. To tap into the dark data, domain experts analyze and extract insights and integrate them into the data repositories. This process can involve out-of-DBMS, ad-hoc analysis, and processing resulting in ETL, engineering effort, and suboptimal performance. While AI systems based on ML models can automate the analysis process, they often further generate context-rich answers. Using multiple sources of truth, for either training the models or in the form of knowledge bases, further exacerbates the problem of consolidating the data of interest. We envision an analytical engine co-optimized with components that enable context-rich analysis. Firstly, as the data from different sources or resulting from model answering cannot be cleaned ahead of time, we propose using online data integration via model-assisted similarity operations. Secondly, we aim for a holistic pipeline cost- and rule-based optimization across relational and model-based operators. Thirdly, with increasingly heterogeneous hardware and equally heterogeneous workloads ranging from traditional relational analytics to generative model inference, we envision a system that just-in-time adapts to the complex analytical query requirements. To solve increasingly complex analytical problems, ML offers attractive solutions that must be combined with traditional analytical processing and benefit from decades of database community research to achieve scalability and performance effortless for the end user.
Extracting phenotypes from clinical text has been shown to be useful for a variety of clinical use cases such as identifying patients with rare diseases. However, reasoning with numerical values remains challenging for phenotyping in clinical text, for example, temperature 102F representing Fever. Current state-of-the-art phenotyping models are able to detect general phenotypes, but perform poorly when they detect phenotypes requiring numerical reasoning. We present a novel unsupervised methodology leveraging external knowledge and contextualized word embeddings from ClinicalBERT for numerical reasoning in a variety of phenotypic contexts. Comparing against unsupervised benchmarks, it shows a substantial performance improvement with absolute gains on generalized Recall and F1 scores up to 79% and 71%, respectively. In the supervised setting, it also surpasses the performance of alternative approaches with absolute gains on generalized Recall and F1 scores up to 70% and 44%, respectively.
General-purpose code generation aims to automatically convert the natural language (NL) description to code snippets in a general-purpose programming language (GPL) like Python. Intrinsically, code generation is a special type of text generation that generates well-formed text, i.e., code. However, existing sequence-to-sequence (Seq2Seq) approaches generate the GPL code neglecting the grammar rules. To this end, in this paper, we make the first attempt to consider grammatical Seq2Seq models for general-purpose code generation and propose CODEP, a grammatical Seq2Seq code generation framework equipped with a Pushdown automaton (PDA) module. In the training stage, CODEP additionally incorporates the state representation and the state prediction task, which leverages PDA states to help CODEP comprehend the parsing process of the PDA module. In the inference stage, CODEP generates well-formed code with the PDA module and the joint prediction of PDA states. Furthermore, the PDA module can be directly applied to Seq2Seq models without training to ensure the grammatical correctness of the generated code. To evaluate the effectiveness of our proposed method, we construct the DPA for the most popular GPL Python and conduct extensive experiments on four benchmark datasets. The experimental results demonstrate the superiority of CODEP compared to the state-of-the-art approaches without pre-training, and the DPA module also achieves significant improvements on the pre-trained models.
Knowledge bases (KB) are an important resource in a number of natural language processing (NLP) and information retrieval (IR) tasks, such as semantic search, automated question-answering etc. They are also useful for researchers trying to gain information from a text. Unfortunately, however, the state-of-the-art in Sanskrit NLP does not yet allow automated construction of knowledge bases due to unavailability or lack of sufficient accuracy of tools and methods. Thus, in this work, we describe our efforts on manual annotation of Sanskrit text for the purpose of knowledge graph (KG) creation. We choose the chapter Dhanyavarga from Bhavaprakashanighantu of the Ayurvedic text Bhavaprakasha for annotation. The constructed knowledge graph contains 410 entities and 764 relationships. Since Bhavaprakashanighantu is a technical glossary text that describes various properties of different substances, we develop an elaborate ontology to capture the semantics of the entity and relationship types present in the text. To query the knowledge graph, we design 31 query templates that cover most of the common question patterns. For both manual annotation and querying, we customize the Sangrahaka framework previously developed by us. The entire system including the dataset is available from https://sanskrit.iitk.ac.in/ayurveda/ . We hope that the knowledge graph that we have created through manual annotation and subsequent curation will help in development and testing of NLP tools in future as well as studying of the Bhavaprakasanighantu text.
Reasoning about actions and change (RAC) is essential to understand and interact with the ever-changing environment. Previous AI research has shown the importance of fundamental and indispensable knowledge of actions, i.e., preconditions and effects. However, traditional methods rely on logical formalization which hinders practical applications. With recent transformer-based language models (LMs), reasoning over text is desirable and seemingly feasible, leading to the question of whether LMs can effectively and efficiently learn to solve RAC problems. We propose four essential RAC tasks as a comprehensive textual benchmark and generate problems in a way that minimizes the influence of other linguistic requirements (e.g., grounding) to focus on RAC. The resulting benchmark, TRAC, encompassing problems of various complexities, facilitates a more granular evaluation of LMs, precisely targeting the structural generalization ability much needed for RAC. Experiments with three high-performing transformers indicates that additional efforts are needed to tackle challenges raised by TRAC.
We introduce KPI-Check, a novel system that automatically identifies and cross-checks semantically equivalent key performance indicators (KPIs), e.g. "revenue" or "total costs", in real-world German financial reports. It combines a financial named entity and relation extraction module with a BERT-based filtering and text pair classification component to extract KPIs from unstructured sentences before linking them to synonymous occurrences in the balance sheet and profit & loss statement. The tool achieves a high matching performance of $73.00$% micro F$_1$ on a hold out test set and is currently being deployed for a globally operating major auditing firm to assist the auditing procedure of financial statements.
This technical report describes the CONE approach for Ego4D Natural Language Queries (NLQ) Challenge in ECCV 2022. We leverage our model CONE, an efficient window-centric COarse-to-fiNE alignment framework. Specifically, CONE dynamically slices the long video into candidate windows via a sliding window approach. Centering at windows, CONE (1) learns the inter-window (coarse-grained) semantic variance through contrastive learning and speeds up inference by pre-filtering the candidate windows relevant to the NL query, and (2) conducts intra-window (fine-grained) candidate moments ranking utilizing the powerful multi-modal alignment ability of the contrastive vision-text pre-trained model EgoVLP. On the blind test set, CONE achieves 15.26 and 9.24 for R1@IoU=0.3 and R1@IoU=0.5, respectively.
Parallel text-to-speech (TTS) models have recently enabled fast and highly-natural speech synthesis. However, such models typically require external alignment models, which are not necessarily optimized for the decoder as they are not jointly trained. In this paper, we propose a differentiable duration method for learning monotonic alignments between input and output sequences. Our method is based on a soft-duration mechanism that optimizes a stochastic process in expectation. Using this differentiable duration method, a direct text to waveform TTS model is introduced to produce raw audio as output instead of performing neural vocoding. Our model learns to perform high-fidelity speech synthesis through a combination of adversarial training and matching the total ground-truth duration. Experimental results show that our model obtains competitive results while enjoying a much simpler training pipeline. Audio samples are available online.
Recent work has demonstrated that pre-trained language models (PLMs) are zero-shot learners. However, most existing zero-shot methods involve heavy human engineering or complicated self-training pipelines, hindering their application to new situations. In this work, we show that zero-shot text classification can be improved simply by clustering texts in the embedding spaces of PLMs. Specifically, we fit the unlabeled texts with a Bayesian Gaussian Mixture Model after initializing cluster positions and shapes using class names. Despite its simplicity, this approach achieves superior or comparable performance on both topic and sentiment classification datasets and outperforms prior works significantly on unbalanced datasets. We further explore the applicability of our clustering approach by evaluating it on 14 datasets with more diverse topics, text lengths, and numbers of classes. Our approach achieves an average of 20% absolute improvement over prompt-based zero-shot learning. Finally, we compare different PLM embedding spaces and find that texts are well-clustered by topics even if the PLM is not explicitly pre-trained to generate meaningful sentence embeddings. This work indicates that PLM embeddings can categorize texts without task-specific fine-tuning, thus providing a new way to analyze and utilize their knowledge and zero-shot learning ability.
An intuitive way to search for images is to use queries composed of an example image and a complementary text. While the first provides rich and implicit context for the search, the latter explicitly calls for new traits, or specifies how some elements of the example image should be changed to retrieve the desired target image. Current approaches typically combine the features of each of the two elements of the query into a single representation, which can then be compared to the ones of the potential target images. Our work aims at shedding new light on the task by looking at it through the prism of two familiar and related frameworks: text-to-image and image-to-image retrieval. Taking inspiration from them, we exploit the specific relation of each query element with the targeted image and derive light-weight attention mechanisms which enable to mediate between the two complementary modalities. We validate our approach on several retrieval benchmarks, querying with images and their associated free-form text modifiers. Our method obtains state-of-the-art results without resorting to side information, multi-level features, heavy pre-training nor large architectures as in previous works.