Contrastive language-image pre-training (CLIP) models have demonstrated considerable success across various vision-language tasks, such as text-to-image retrieval, where the model is required to effectively process natural language input to produce an accurate visual output. However, current models still face limitations in dealing with linguistic variations in input queries, such as paraphrases, making it challenging to handle a broad range of user queries in real-world applications. In this study, we introduce a straightforward fine-tuning approach to enhance the representations of CLIP models for paraphrases. Our approach involves a two-step paraphrase generation process, where we automatically create two categories of paraphrases from web-scale image captions by leveraging large language models. Subsequently, we fine-tune the CLIP text encoder using these generated paraphrases while freezing the image encoder. Our resulting model, which we call ParaCLIP, exhibits significant improvements over baseline CLIP models across various tasks, including paraphrased retrieval (with rank similarity scores improved by up to 7.6% and 9.6%), Visual Genome Relation and Attribution, as well as seven semantic textual similarity tasks.

While contextualized word embeddings have been a de-facto standard, learning contextualized phrase embeddings is less explored and being hindered by the lack of a human-annotated benchmark that tests machine understanding of phrase semantics given a context sentence or paragraph (instead of phrases alone). To fill this gap, we propose PiC—a dataset of ∼28K of noun phrases accompanied by their contextual Wikipedia pages and a suite of three tasks for training and evaluating phrase embeddings. Training on PiC improves ranking-models’ accuracy and remarkably pushes span selection (SS) models (i.e., predicting the start and end index of the target phrase) near human accuracy, which is 95% Exact Match (EM) on semantic search given a query phrase and a passage. Interestingly, we find evidence that such impressive performance is because the SS models learn to better capture the common meaning of a phrase regardless of its actual context. SotA models perform poorly in distinguishing two senses of the same phrase in two contexts (∼60% EM) and in estimating the similarity between two different phrases in the same context (∼70% EM).

Vulnerability to lexical perturbation is a critical weakness of automatic evaluation metrics for image captioning. This paper proposes Perturbation Robust Multi-Lingual CLIPScore(PR-MCS), which exhibits robustness to such perturbations, as a novel reference-free image captioning metric applicable to multiple languages. To achieve perturbation robustness, we fine-tune the text encoder of CLIP with our language-agnostic method to distinguish the perturbed text from the original text. To verify the robustness of PR-MCS, we introduce a new fine-grained evaluation dataset consisting of detailed captions, critical objects, and the relationships between the objects for 3,000 images in five languages. In our experiments, PR-MCS significantly outperforms baseline metrics in capturing lexical noise of all various perturbation types in all five languages, while maintaining a strong correlation with human judgments.

Most weakly supervised named entity recognition (NER) models rely on domain-specific dictionaries provided by experts. This approach is infeasible in many domains where dictionaries do not exist. While a phrase retrieval model was used to construct pseudo-dictionaries with entities retrieved from Wikipedia automatically in a recent study, these dictionaries often have limited coverage because the retriever is likely to retrieve popular entities rather than rare ones. In this study, we present a novel framework, HighGEN, that generates NER datasets with high-coverage pseudo-dictionaries. Specifically, we create entity-rich dictionaries with a novel search method, called phrase embedding search, which encourages the retriever to search a space densely populated with various entities. In addition, we use a new verification process based on the embedding distance between candidate entity mentions and entity types to reduce the false-positive noise in weak labels generated by high-coverage dictionaries. We demonstrate that HighGEN outperforms the previous best model by an average F1 score of 4.7 across five NER benchmark datasets.

With the ubiquitous use of online meeting platforms and robust automatic speech recognition systems, meeting transcripts have emerged as a promising domain for natural language tasks. Most recent works on meeting transcripts primarily focus on summarization and extraction of action items. However, meeting discussions also have a useful question-answering (QA) component, crucial to understanding the discourse or meeting content, and can be used to build interactive interfaces on top of long transcripts. Hence, in this work, we leverage this inherent QA component of meeting discussions and introduce MeetingQA, an extractive QA dataset comprising of questions asked by meeting participants and corresponding responses. As a result, questions can be open-ended and actively seek discussions, while the answers can be multi-span and distributed across multiple speakers. Our comprehensive empirical study of several robust baselines including long-context language models and recent instruction-tuned models reveals that models perform poorly on this task (F1 = 57.3) and severely lag behind human performance (F1 = 84.6), thus presenting a challenging new task for the community to improve upon.

Despite the recent advancements in abstractive summarization systems leveraged from large-scale datasets and pre-trained language models, the factual correctness of the summary is still insufficient. One line of trials to mitigate this problem is to include a post-editing process that can detect and correct factual errors in the summary. In building such a system, it is strongly required that 1) the process has a high success rate and interpretability and 2) it has a fast running time. Previous approaches focus on the regeneration of the summary, resulting in low interpretability and high computing resources. In this paper, we propose an efficient factual error correction system RFEC based on entity retrieval. RFEC first retrieves the evidence sentences from the original document by comparing the sentences with the target summary to reduce the length of the text to analyze. Next, RFEC detects entity-level errors in the summaries using the evidence sentences and substitutes the wrong entities with the accurate entities from the evidence sentences. Experimental results show that our proposed error correction system shows more competitive performance than baseline methods in correcting factual errors with a much faster speed.

Individuals, educational institutions, and businesses are prolific at generating instructional video content such as “how-to” and tutorial guides. While significant progress has been made in basic video understanding tasks, identifying procedural intent within these instructional videos is a challenging and important task that remains unexplored but essential to video summarization, search, and recommendations. This paper introduces the problem of instructional intent identification and extraction from software instructional livestreams. We construct and present a new multimodal dataset consisting of software instructional livestreams and containing manual annotations for both detailed and abstract procedural intent that enable training and evaluation of joint video and text understanding models. We then introduce a multimodal cascaded cross-attention model to efficiently combine the weaker and noisier video signal with the more discriminative text signal. Our experiments show that our proposed model brings significant gains compared to strong baselines, including large-scale pretrained multimodal models. Our analysis further identifies that the task benefits from spatial as well as motion features extracted from videos, and provides insight on how the video signal is preferentially used for intent discovery. We also show that current models struggle to comprehend the nature of abstract intents, revealing important gaps in multimodal understanding and paving the way for future work.

Modern image captioning models are usually trained with text similarity objectives. However, since reference captions in public datasets often describe the most salient common objects, models trained with the text similarity objectives tend to ignore specific and detailed aspects of an image that distinguish it from others. Towards more descriptive and distinctive caption generation, we propose to use CLIP, a multimodal encoder trained on huge image-text pairs from the web, to calculate multi-modal similarity and use it as a reward function. We also propose a simple finetuning strategy of CLIP text encoder to improve grammar that does not require extra text annotation. This completely eliminates the need for reference captions during the reward computation. To comprehensively evaluate descriptive captions, we introduce FineCapEval, a new dataset for caption evaluation with fine-grained criteria: overall, background, object, relations. In our experiments on text-to-image retrieval and FineCapEval, the proposed CLIP-guided model generates more distinctive captions than the CIDEroptimized model. We also show that our unsupervised grammar finetuning of the CLIP text encoder alleviates the degeneration problem of the naive CLIP reward. Lastly, we show human analysis where the annotators strongly prefer CLIP reward to CIDEr and MLE objectives on diverse criteria.

Recent named entity recognition (NER) models often rely on human-annotated datasets requiring the vast engagement of professional knowledge on the target domain and entities. This work introduces an ask-to-generate approach, which automatically generates NER datasets by asking simple natural language questions to an open-domain question answering system (e.g., “Which disease?”). Despite using fewer training resources, our models solely trained on the generated datasets largely outperform strong low-resource models by 19.5 F1 score across six popular NER benchmarks. Our models also show competitive performance with rich-resource models that additionally leverage in-domain dictionaries provided by domain experts. In few-shot NER, we outperform the previous best model by 5.2 F1 score on three benchmarks and achieve new state-of-the-art performance.

This study investigates how fake news use the thumbnail image for a news article. We aim at capturing the degree of semantic incongruity between news text and image by using the pretrained CLIP representation. Motivated by the stylistic distinctiveness in fake news text, we examine whether fake news tends to use an irrelevant image to the news content. Results show that fake news tends to have a high degree of semantic incongruity than general news. We further attempt to detect such image-text incongruity by training classification models on a newly generated dataset. A manual evaluation suggests our method can find news articles of which the thumbnail image is semantically irrelevant to news text with an accuracy of 0.8. We also release a new dataset of image and news text pairs with the incongruity label, facilitating future studies on the direction.

Domain-specific documents cover terminologies and specialized knowledge. This has been the main challenge of domain-specific document retrieval systems. Previous approaches propose domain-adaptation and transfer learning methods to alleviate this problem. However, these approaches still follow the same document representation method in previous approaches; a document is embedded into a single vector. In this study, we propose VKGDR. VKGDR represents a given corpus into a graph of entities and their relations (known as a virtual knowledge graph) and computes the relevance between queries and documents based on the graph representation. We conduct three experiments 1) domain-specific document retrieval, 2) comparison of our virtual knowledge graph construction method with previous approaches, and 3) ablation study on each component of our virtual knowledge graph. From the results, we see that unsupervised VKGDR outperforms baselines in a zero-shot setting and even outperforms fully-supervised bi-encoder. We also verify that our virtual knowledge graph construction method results in better retrieval performance than previous approaches.

Current medical question answering systems have difficulty processing long, detailed and informally worded questions submitted by patients, called Consumer Health Questions (CHQs). To address this issue, we introduce a medical question understanding and answering system with knowledge grounding and semantic self-supervision. Our system is a pipeline that first summarizes a long, medical, user-written question, using a supervised summarization loss. Then, our system performs a two-step retrieval to return answers. The system first matches the summarized user question with an FAQ from a trusted medical knowledge base, and then retrieves a fixed number of relevant sentences from the corresponding answer document. In the absence of labels for question matching or answer relevance, we design 3 novel, self-supervised and semantically-guided losses. We evaluate our model against two strong retrieval-based question answering baselines. Evaluators ask their own questions and rate the answers retrieved by our baselines and own system according to their relevance. They find that our system retrieves more relevant answers, while achieving speeds 20 times faster. Our self-supervised losses also help the summarizer achieve higher scores in ROUGE, as well as in human evaluation metrics.

Acronym extraction is the task of identifying acronyms and their expanded forms in texts that is necessary for various NLP applications. Despite major progress for this task in recent years, one limitation of existing AE research is that they are limited to the English language and certain domains (i.e., scientific and biomedical). Challenges of AE in other languages and domains are mainly unexplored. As such, lacking annotated datasets in multiple languages and domains has been a major issue to prevent research in this direction. To address this limitation, we propose a new dataset for multilingual and multi-domain AE. Specifically, 27,200 sentences in 6 different languages and 2 new domains, i.e., legal and scientific, are manually annotated for AE. Our experiments on the dataset show that AE in different languages and learning settings has unique challenges, emphasizing the necessity of further research on multilingual and multi-domain AE.

The prevalent use of offensive content in social media has become an important reason for concern for online platforms (customer service chat-boxes, social media platforms, etc). Classifying offensive and hate-speech content in online settings is an essential task in many applications that needs to be addressed accordingly. However, online text from online platforms can contain code-switching, a combination of more than one language. The non-availability of labeled code-switched data for low-resourced code-switching combinations adds difficulty to this problem. To overcome this, we release a real-world dataset containing around 10k samples for testing for three language combinations en-fr, en-es, and en-de, and a synthetic code-switched textual dataset containing ~30k samples for training In this paper, we describe the process for gathering the human-generated data and our algorithm for creating synthetic code-switched offensive content data. We also introduce the results of a keyword classification baseline and a multi-lingual transformer-based classification model.

Keyphrase Prediction (KP) is an established NLP task, aiming to yield representative phrases to summarize the main content of a given document. Despite major progress in recent years, existing works on KP have mainly focused on formal texts such as scientific papers or weblogs. The challenges of KP in informal-text domains are not yet fully studied. To this end, this work studies new challenges of KP in transcripts of videos, an understudied domain for KP that involves informal texts and non-cohesive presentation styles. A bottleneck for KP research in this domain involves the lack of high-quality and large-scale annotated data that hinders the development of advanced KP models. To address this issue, we introduce a large-scale manually-annotated KP dataset in the domain of live-stream video transcripts obtained by automatic speech recognition tools. Concretely, transcripts of 500+ hours of videos streamed on the behance.net platform are manually labeled with important keyphrases. Our analysis of the dataset reveals the challenging nature of KP in transcripts. Moreover, for the first time in KP, we demonstrate the idea of improving KP for long documents (i.e., transcripts) by feeding models with paragraph-level keyphrases, i.e., hierarchical extraction. To foster future research, we will publicly release the dataset and code.

In the automatic evaluation of generative question answering (GenQA) systems, it is difficult to assess the correctness of generated answers due to the free-form of the answer. Especially, widely used n-gram similarity metrics often fail to discriminate the incorrect answers since they equally consider all of the tokens. To alleviate this problem, we propose KPQA metric, a new metric for evaluating the correctness of GenQA. Specifically, our new metric assigns different weights to each token via keyphrase prediction, thereby judging whether a generated answer sentence captures the key meaning of the reference answer. To evaluate our metric, we create high-quality human judgments of correctness on two GenQA datasets. Using our human-evaluation datasets, we show that our proposed metric has a significantly higher correlation with human judgments than existing metrics in various datasets. Code for KPQA-metric will be available at https://github.com/hwanheelee1993/KPQA.

In this work, we focus on a more challenging few-shot intent detection scenario where many intents are fine-grained and semantically similar. We present a simple yet effective few-shot intent detection schema via contrastive pre-training and fine-tuning. Specifically, we first conduct self-supervised contrastive pre-training on collected intent datasets, which implicitly learns to discriminate semantically similar utterances without using any labels. We then perform few-shot intent detection together with supervised contrastive learning, which explicitly pulls utterances from the same intent closer and pushes utterances across different intents farther. Experimental results show that our proposed method achieves state-of-the-art performance on three challenging intent detection datasets under 5-shot and 10-shot settings.

Users of medical question answering systems often submit long and detailed questions, making it hard to achieve high recall in answer retrieval. To alleviate this problem, we propose a novel Multi-Task Learning (MTL) method with data augmentation for medical question understanding. We first establish an equivalence between the tasks of question summarization and Recognizing Question Entailment (RQE) using their definitions in the medical domain. Based on this equivalence, we propose a data augmentation algorithm to use just one dataset to optimize for both tasks, with a weighted MTL loss. We introduce gradually soft parameter-sharing: a constraint for decoder parameters to be close, that is gradually loosened as we move to the highest layer. We show through ablation studies that our proposed novelties improve performance. Our method outperforms existing MTL methods across 4 datasets of medical question pairs, in ROUGE scores, RQE accuracy and human evaluation. Finally, we show that our method fares better than single-task learning under 4 low-resource settings.

Despite the success of various text generation metrics such as BERTScore, it is still difficult to evaluate the image captions without enough reference captions due to the diversity of the descriptions. In this paper, we introduce a new metric UMIC, an Unreferenced Metric for Image Captioning which does not require reference captions to evaluate image captions. Based on Vision-and-Language BERT, we train UMIC to discriminate negative captions via contrastive learning. Also, we observe critical problems of the previous benchmark dataset (i.e., human annotations) on image captioning metric, and introduce a new collection of human annotations on the generated captions. We validate UMIC on four datasets, including our new dataset, and show that UMIC has a higher correlation than all previous metrics that require multiple references.

In this paper, we describe our approach to question summarization and multi-answer summarization in the context of the 2021 MEDIQA shared task (Ben Abacha et al., 2021). We propose two kinds of transfer learning for the abstractive summarization of medical questions. First, we train on HealthCareMagic, a large question summarization dataset collected from an online healthcare service platform. Second, we leverage the ability of the BART encoder-decoder architecture to model both generation and classification tasks to train on the task of Recognizing Question Entailment (RQE) in the medical domain. We show that both transfer learning methods combined achieve the highest ROUGE scores. Finally, we cast the question-driven extractive summarization of multiple relevant answer documents as an Answer Sentence Selection (AS2) problem. We show how we can preprocess the MEDIQA-AnS dataset such that it can be trained in an AS2 setting. Our AS2 model is able to generate extractive summaries achieving high ROUGE scores.

In this paper we propose QACE, a new metric based on Question Answering for Caption Evaluation to evaluate image captioning based on Question Generation(QG) and Question Answering(QA) systems. QACE generates questions on the evaluated caption and check its content by asking the questions on either the reference caption or the source image. We first develop QACE_Ref that compares the answers of the evaluated caption to its reference, and report competitive results with the state-of-the-art metrics. To go further, we propose QACE_Img, that asks the questions directly on the image, instead of reference. A Visual-QA system is necessary for QACE_Img. Unfortunately, the standard VQA models are actually framed a classification among only few thousands categories. Instead, we propose Visual-T5, an abstractive VQA system. The resulting metric, QACE_Img is multi-modal, reference-less and explainable. Our experiments show that QACE_Img compares favorably w.r.t. other reference-less metrics.

In this study, we propose a novel graph neural network called propagate-selector (PS), which propagates information over sentences to understand information that cannot be inferred when considering sentences in isolation. First, we design a graph structure in which each node represents an individual sentence, and some pairs of nodes are selectively connected based on the text structure. Then, we develop an iterative attentive aggregation and a skip-combine method in which a node interacts with its neighborhood nodes to accumulate the necessary information. To evaluate the performance of the proposed approaches, we conduct experiments with the standard HotpotQA dataset. The empirical results demonstrate the superiority of our proposed approach, which obtains the best performances, compared to the widely used answer-selection models that do not consider the intersentential relationship.

In this paper, we propose an evaluation metric for image captioning systems using both image and text information. Unlike the previous methods that rely on textual representations in evaluating the caption, our approach uses visiolinguistic representations. The proposed method generates image-conditioned embeddings for each token using ViLBERT from both generated and reference texts. Then, these contextual embeddings from each of the two sentence-pair are compared to compute the similarity score. Experimental results on three benchmark datasets show that our method correlates significantly better with human judgments than all existing metrics.

Even though BERT has achieved successful performance improvements in various supervised learning tasks, BERT is still limited by repetitive inferences on unsupervised tasks for the computation of contextual language representations. To resolve this limitation, we propose a novel deep bidirectional language model called a Transformer-based Text Autoencoder (T-TA). The T-TA computes contextual language representations without repetition and displays the benefits of a deep bidirectional architecture, such as that of BERT. In computation time experiments in a CPU environment, the proposed T-TA performs over six times faster than the BERT-like model on a reranking task and twelve times faster on a semantic similarity task. Furthermore, the T-TA shows competitive or even better accuracies than those of BERT on the above tasks. Code is available at https://github.com/joongbo/tta.

While deep learning techniques have shown promising results in many natural language processing (NLP) tasks, it has not been widely applied to the clinical domain. The lack of large datasets and the pervasive use of domain-specific language (i.e. abbreviations and acronyms) in the clinical domain causes slower progress in NLP tasks than that of the general NLP tasks. To fill this gap, we employ word/subword-level based models that adopt large-scale data-driven methods such as pre-trained language models and transfer learning in analyzing text for the clinical domain. Empirical results demonstrate the superiority of the proposed methods by achieving 90.6% accuracy in medical domain natural language inference task. Furthermore, we inspect the independent strengths of the proposed approaches in quantitative and qualitative manners. This analysis will help researchers to select necessary components in building models for the medical domain.

In this paper, we propose a novel end-to-end neural architecture for ranking candidate answers, that adapts a hierarchical recurrent neural network and a latent topic clustering module. With our proposed model, a text is encoded to a vector representation from an word-level to a chunk-level to effectively capture the entire meaning. In particular, by adapting the hierarchical structure, our model shows very small performance degradations in longer text comprehension while other state-of-the-art recurrent neural network models suffer from it. Additionally, the latent topic clustering module extracts semantic information from target samples. This clustering module is useful for any text related tasks by allowing each data sample to find its nearest topic cluster, thus helping the neural network model analyze the entire data. We evaluate our models on the Ubuntu Dialogue Corpus and consumer electronic domain question answering dataset, which is related to Samsung products. The proposed model shows state-of-the-art results for ranking question-answer pairs.

The context-dependent nature of online aggression makes annotating large collections of data extremely difficult. Previously studied datasets in abusive language detection have been insufficient in size to efficiently train deep learning models. Recently, Hate and Abusive Speech on Twitter, a dataset much greater in size and reliability, has been released. However, this dataset has not been comprehensively studied to its potential. In this paper, we conduct the first comparative study of various learning models on Hate and Abusive Speech on Twitter, and discuss the possibility of using additional features and context data for improvements. Experimental results show that bidirectional GRU networks trained on word-level features, with Latent Topic Clustering modules, is the most accurate model scoring 0.805 F1.