Detecting logical fallacies in texts can help users spot argument flaws, but automating this detection is not easy. Manually annotating fallacies in large-scale, real-world text data to create datasets for developing and validating detection models is costly. This paper introduces CoCoLoFa, the largest known logical fallacy dataset, containing 7,706 comments for 648 news articles, with each comment labeled for fallacy presence and type. We recruited 143 crowd workers to write comments embodying specific fallacy types (e.g., slippery slope) in response to news articles. Recognizing the complexity of this writing task, we built an LLM-powered assistant into the workers’ interface to aid in drafting and refining their comments. Experts rated the writing quality and labeling validity of CoCoLoFa as high and reliable. BERT-based models fine-tuned using CoCoLoFa achieved the highest fallacy detection (F1=0.86) and classification (F1=0.87) performance on its test set, outperforming the state-of-the-art LLMs. Our work shows that combining crowdsourcing and LLMs enables us to more effectively construct datasets for complex linguistic phenomena that crowd workers find challenging to produce on their own.

The rapid growth of scientific publications, particularly during the COVID-19 pandemic, emphasizes the need for tools to help researchers efficiently comprehend the latest advancements. One essential part of understanding scientific literature is research aspect classification, which categorizes sentences in abstracts to Background, Purpose, Method, and Finding. In this study, we investigate the impact of different datasets on model performance for the crowd-annotated CODA-19 research aspect classification task. Specifically, we explore the potential benefits of using the large, automatically curated PubMed 200K RCT dataset and evaluate the effectiveness of large language models (LLMs), such as LLaMA, GPT-3, ChatGPT, and GPT-4. Our results indicate that using the PubMed 200K RCT dataset does not improve performance for the CODA-19 task. We also observe that while GPT-4 performs well, it does not outperform the SciBERT model fine-tuned on the CODA-19 dataset, emphasizing the importance of a dedicated and task-aligned datasets dataset for the target task.

There is growing interest in systems that generate captions for scientific figures. However, assessing these systems’ output poses a significant challenge. Human evaluation requires academic expertise and is costly, while automatic evaluation depends on often low-quality author-written captions. This paper investigates using large language models (LLMs) as a cost-effective, reference-free method for evaluating figure captions. We first constructed SCICAP-EVAL, a human evaluation dataset that contains human judgments for 3,600 scientific figure captions, both original and machine-made, for 600 arXiv figures. We then prompted LLMs like GPT-4 and GPT-3 to score (1-6) each caption based on its potential to aid reader understanding, given relevant context such as figure-mentioning paragraphs. Results show that GPT-4, used as a zero-shot evaluator, outperformed all other models and even surpassed assessments made by computer science undergraduates, achieving a Kendall correlation score of 0.401 with Ph.D. students’ rankings.

This work introduces a novel task, location-aware visual question generation (LocaVQG), which aims to generate engaging questions from data relevant to a particular geographical location. Specifically, we represent such location-aware information with surrounding images and a GPS coordinate. To tackle this task, we present a dataset generation pipeline that leverages GPT-4 to produce diverse and sophisticated questions. Then, we aim to learn a lightweight model that can address the LocaVQG task and fit on an edge device, such as a mobile phone. To this end, we propose a method which can reliably generate engaging questions from location-aware information. Our proposed method outperforms baselines regarding human evaluation (e.g., engagement, grounding, coherence) and automatic evaluation metrics (e.g., BERTScore, ROUGE-2). Moreover, we conduct extensive ablation studies to justify our proposed techniques for both generating the dataset and solving the task.

Little attention is placed on analyzing nationality bias in language models, especially when nationality is highly used as a factor in increasing the performance of social NLP models. This paper examines how a text generation model, GPT-2, accentuates pre-existing societal biases about country-based demonyms. We generate stories using GPT-2 for various nationalities and use sensitivity analysis to explore how the number of internet users and the country’s economic status impacts the sentiment of the stories. To reduce the propagation of biases through large language models (LLM), we explore the debiasing method of adversarial triggering. Our results show that GPT-2 demonstrates significant bias against countries with lower internet users, and adversarial triggering effectively reduces the same.

Good figure captions help paper readers understand complex scientific figures. Unfortunately, even published papers often have poorly written captions. Automatic caption generation could aid paper writers by providing good starting captions that can be refined for better quality. Prior work often treated figure caption generation as a vision-to-language task. In this paper, we show that it can be more effectively tackled as a text summarization task in scientific documents. We fine-tuned PEGASUS, a pre-trained abstractive summarization model, to specifically summarize figure-referencing paragraphs (e.g., “Figure 3 shows...”) into figure captions. Experiments on large-scale arXiv figures show that our method outperforms prior vision methods in both automatic and human evaluations. We further conducted an in-depth investigation focused on two key challenges: (i) the common presence of low-quality author-written captions and (ii) the lack of clear standards for good captions. Our code and data are available at: https://github.com/Crowd-AI-Lab/Generating-Figure-Captions-as-a-Text-Summarization-Task.

Visual storytelling (VIST) is a typical vision and language task that has seen extensive development in the natural language generation research domain. However, it remains unclear whether conventional automatic evaluation metrics for text generation are applicable on VIST. In this paper, we present the VHED (VIST Human Evaluation Data) dataset, which first re-purposes human evaluation results for automatic evaluation; hence we develop Vrank (VIST Ranker), a novel reference-free VIST metric for story evaluation. We first show that the results from commonly adopted automatic metrics for text generation have little correlation with those obtained from human evaluation, which motivates us to directly utilize human evaluation results to learn the automatic evaluation model. In the experiments, we evaluate the generated texts to predict story ranks using our model as well as other reference-based and reference-free metrics. Results show that Vrank prediction is significantly more aligned to human evaluation than other metrics with almost 30% higher accuracy when ranking story pairs. Moreover, we demonstrate that only Vrank shows human-like behavior in its strong ability to find better stories when the quality gap between two stories is high. Finally, we show the superiority of Vrank by its generalizability to pure textual stories, and conclude that this reuse of human evaluation results puts Vrank in a strong position for continued future advances.

Existing self-explaining models typically favor extracting the shortest possible rationales — snippets of an input text “responsible for” corresponding output — to explain the model prediction, with the assumption that shorter rationales are more intuitive to humans. However, this assumption has yet to be validated. Is the shortest rationale indeed the most human-understandable? To answer this question, we design a self-explaining model, LimitedInk, which allows users to extract rationales at any target length. Compared to existing baselines, LimitedInk achieves compatible end-task performance and human-annotated rationale agreement, making it a suitable representation of the recent class of self-explaining models. We use LimitedInk to conduct a user study on the impact of rationale length, where we ask human judges to predict the sentiment label of documents based only on LimitedInk-generated rationales with different lengths. We show rationales that are too short do not help humans predict labels better than randomly masked text, suggesting the need for more careful design of the best human rationales.

Generating engaging content has drawn much recent attention in the NLP community. Asking questions is a natural way to respond to photos and promote awareness. However, most answers to questions in traditional question-answering (QA) datasets are factoids, which reduce individuals’ willingness to answer. Furthermore, traditional visual question generation (VQG) confines the source data for question generation to single images, resulting in a limited ability to comprehend time-series information of the underlying event. In this paper, we propose generating engaging questions from multiple images. We present MVQG, a new dataset, and establish a series of baselines, including both end-to-end and dual-stage architectures. Results show that building stories behind the image sequence enables models togenerate engaging questions, which confirms our assumption that people typically construct a picture of the event in their minds before asking questions. These results open up an exciting challenge for visual-and-language models to implicitly construct a story behind a series of photos to allow for creativity and experience sharing and hence draw attention to downstream applications.

Atomic clauses are fundamental text units for understanding complex sentences. Identifying the atomic sentences within complex sentences is important for applications such as summarization, argument mining, discourse analysis, discourse parsing, and question answering. Previous work mainly relies on rule-based methods dependent on parsing. We propose a new task to decompose each complex sentence into simple sentences derived from the tensed clauses in the source, and a novel problem formulation as a graph edit task. Our neural model learns to Accept, Break, Copy or Drop elements of a graph that combines word adjacency and grammatical dependencies. The full processing pipeline includes modules for graph construction, graph editing, and sentence generation from the output graph. We introduce DeSSE, a new dataset designed to train and evaluate complex sentence decomposition, and MinWiki, a subset of MinWikiSplit. ABCD achieves comparable performance as two parsing baselines on MinWiki. On DeSSE, which has a more even balance of complex sentence types, our model achieves higher accuracy on the number of atomic sentences than an encoder-decoder baseline. Results include a detailed error analysis.

In visual storytelling, a short story is generated based on a given image sequence. Despite years of work, most visual storytelling models remain limited in terms of the generated stories’ fixed length: most models produce stories with exactly five sentences because five-sentence stories dominate the training data. The fix-length stories carry limited details and provide ambiguous textual information to the readers. Therefore, we propose to “stretch” the stories, which create the potential to present in-depth visual details. This paper presents Stretch-VST, a visual storytelling framework that enables the generation of prolonged stories by adding appropriate knowledge, which is selected by the proposed scoring function. We propose a length-controlled Transformer to generate long stories. This model introduces novel positional encoding methods to maintain story quality with lengthy inputs. Experiments confirm that long stories are generated without deteriorating the quality. The human evaluation further shows that Stretch-VST can provide better focus and detail when stories are prolonged compared to state of the art. We create a webpage to demonstrate our prolonged capability.

This paper introduces Semantic Frame Forecast, a task that predicts the semantic frames that will occur in the next 10, 100, or even 1,000 sentences in a running story. Prior work focused on predicting the immediate future of a story, such as one to a few sentences ahead. However, when novelists write long stories, generating a few sentences is not enough to help them gain high-level insight to develop the follow-up story. In this paper, we formulate a long story as a sequence of “story blocks,” where each block contains a fixed number of sentences (e.g., 10, 100, or 200). This formulation allows us to predict the follow-up story arc beyond the scope of a few sentences. We represent a story block using the term frequencies (TF) of semantic frames in it, normalized by each frame’s inverse document frequency (IDF). We conduct semantic frame forecast experiments on 4,794 books from the Bookcorpus and 7,962 scientific abstracts from CODA-19, with block sizes ranging from 5 to 1,000 sentences. The results show that automated models can forecast the follow-up story blocks better than the random, prior, and replay baselines, indicating the feasibility of the task. We also learn that the models using the frame representation as features outperform all the existing approaches when the block size is over 150 sentences. The human evaluation also shows that the proposed frame representation, when visualized as word clouds, is comprehensible, representative, and specific to humans.

Semantic representation that supports the choice of an appropriate connective between pairs of clauses inherently addresses discourse coherence, which is important for tasks such as narrative understanding, argumentation, and discourse parsing. We propose a novel clause embedding method that applies graph learning to a data structure we refer to as a dependency-anchor graph. The dependency anchor graph incorporates two kinds of syntactic information, constituency structure, and dependency relations, to highlight the subject and verb phrase relation. This enhances coherence-related aspects of representation. We design a neural model to learn a semantic representation for clauses from graph convolution over latent representations of the subject and verb phrase. We evaluate our method on two new datasets: a subset of a large corpus where the source texts are published novels, and a new dataset collected from students’ essays. The results demonstrate a significant improvement over tree-based models, confirming the importance of emphasizing the subject and verb phrase. The performance gap between the two datasets illustrates the challenges of analyzing student’s written text, plus a potential evaluation task for coherence modeling and an application for suggesting revisions to students.

Researchers use figures to communicate rich, complex information in scientific papers. The captions of these figures are critical to conveying effective messages. However, low-quality figure captions commonly occur in scientific articles and may decrease understanding. In this paper, we propose an end-to-end neural framework to automatically generate informative, high-quality captions for scientific figures. To this end, we introduce SCICAP, a large-scale figure-caption dataset based on computer science arXiv papers published between 2010 and 2020. After pre-processing – including figure-type classification, sub-figure identification, text normalization, and caption text selection – SCICAP contained more than two million figures extracted from over 290,000 papers. We then established baseline models that caption graph plots, the dominant (19.2%) figure type. The experimental results showed both opportunities and steep challenges of generating captions for scientific figures.

The sheer volume of financial statements makes it difficult for humans to access and analyze a business’s financials. Robust numerical reasoning likewise faces unique challenges in this domain. In this work, we focus on answering deep questions over financial data, aiming to automate the analysis of a large corpus of financial documents. In contrast to existing tasks on general domain, the finance domain includes complex numerical reasoning and understanding of heterogeneous representations. To facilitate analytical progress, we propose a new large-scale dataset, FinQA, with Question-Answering pairs over Financial reports, written by financial experts. We also annotate the gold reasoning programs to ensure full explainability. We further introduce baselines and conduct comprehensive experiments in our dataset. The results demonstrate that popular, large, pre-trained models fall far short of expert humans in acquiring finance knowledge and in complex multi-step numerical reasoning on that knowledge. Our dataset – the first of its kind – should therefore enable significant, new community research into complex application domains. The dataset and code are publicly available at https://github.com/czyssrs/FinQA.

This paper introduces CODA-19, a human-annotated dataset that codes the Background, Purpose, Method, Finding/Contribution, and Other sections of 10,966 English abstracts in the COVID-19 Open Research Dataset. CODA-19 was created by 248 crowd workers from Amazon Mechanical Turk within 10 days, and achieved labeling quality comparable to that of experts. Each abstract was annotated by nine different workers, and the final labels were acquired by majority vote. The inter-annotator agreement (Cohen’s kappa) between the crowd and the biomedical expert (0.741) is comparable to inter-expert agreement (0.788). CODA-19’s labels have an accuracy of 82.2% when compared to the biomedical expert’s labels, while the accuracy between experts was 85.0%. Reliable human annotations help scientists access and integrate the rapidly accelerating coronavirus literature, and also serve as the battery of AI/NLP research, but obtaining expert annotations can be slow. We demonstrated that a non-expert crowd can be rapidly employed at scale to join the fight against COVID-19.

Many English-as-a-second language learners have trouble using near-synonym words (e.g., small vs.little; briefly vs.shortly) correctly, and often look for example sentences to learn how two nearly synonymous terms differ. Prior work uses hand-crafted scores to recommend sentences but has difficulty in adopting such scores to all the near-synonyms as near-synonyms differ in various ways. We notice that the helpfulness of the learning material would reflect on the learners’ performance. Thus, we propose the inference-based learner-like agent to mimic learner behavior and identify good learning materials by examining the agent’s performance. To enable the agent to behave like a learner, we leverage entailment modeling’s capability of inferring answers from the provided materials. Experimental results show that the proposed agent is equipped with good learner-like behavior to achieve the best performance in both fill-in-the-blank (FITB) and good example sentence selection tasks. We further conduct a classroom user study with college ESL learners. The results of the user study show that the proposed agent can find out example sentences that help students learn more easily and efficiently. Compared to other models, the proposed agent improves the score of more than 17% of students after learning.

We introduce the first dataset for human edits of machine-generated visual stories and explore how these collected edits may be used for the visual story post-editing task. The dataset ,VIST-Edit, includes 14,905 human-edited versions of 2,981 machine-generated visual stories. The stories were generated by two state-of-the-art visual storytelling models, each aligned to 5 human-edited versions. We establish baselines for the task, showing how a relatively small set of human edits can be leveraged to boost the performance of large visual storytelling models. We also discuss the weak correlation between automatic evaluation scores and human ratings, motivating the need for new automatic metrics.

We present MoodSwipe, a soft keyboard that suggests text messages given the user-specified emotions utilizing the real dialog data. The aim of MoodSwipe is to create a convenient user interface to enjoy the technology of emotion classification and text suggestion, and at the same time to collect labeled data automatically for developing more advanced technologies. While users select the MoodSwipe keyboard, they can type as usual but sense the emotion conveyed by their text and receive suggestions for their message as a benefit. In MoodSwipe, the detected emotions serve as the medium for suggested texts, where viewing the latter is the incentive to correcting the former. We conduct several experiments to show the superiority of the emotion classification models trained on the dialog data, and further to verify good emotion cues are important context for text suggestion.

Instant messaging and push notifications play important roles in modern digital life. To enable robust sense-making and rich context awareness in computer mediated communications, we introduce EmotionPush, a system that automatically conveys the emotion of received text with a colored push notification on mobile devices. EmotionPush is powered by state-of-the-art emotion classifiers and is deployed for Facebook Messenger clients on Android. The study showed that the system is able to help users prioritize interactions.

In this paper, we base on the syntactic structural Chinese Treebank corpus, construct the Chinese Opinon Treebank for the research of opinion analysis. We introduce the tagging scheme and develop a tagging tool for constructing this corpus. Annotated samples are described. Information including opinions (yes or no), their polarities (positive, neutral or negative), types (expression, status, or action), is defined and annotated. In addition, five structure trios are introduced according to the linguistic relations between two Chinese words. Four of them that are possibly related to opinions are also annotated in the constructed corpus to provide the linguistic cues. The number of opinion sentences together with the number of their polarities, opinion types, and trio types are calculated. These statistics are compared and discussed. To know the quality of the annotations in this corpus, the kappa values of the annotations are calculated. The substantial agreement between annotations ensures the applicability and reliability of the constructed corpus.

This paper presented an overview of Chinese bi-character words’ morphological types, and proposed a set of features for machine learning approaches to predict these types based on composite characters’ information. First, eight morphological types were defined, and 6,500 Chinese bi-character words were annotated with these types. After pre-processing, 6,178 words were selected to construct a corpus named Reduced Set. We analyzed Reduced Set and conducted the inter-annotator agreement test. The average kappa value of 0.67 indicates a substantial agreement. Second, Bi-character words’ morphological types are considered strongly related with the composite characters’ parts of speech in this paper, so we proposed a set of features which can simply be extracted from dictionaries to indicate the characters’ “tendency” of parts of speech. Finally, we used these features and adopted three machine learning algorithms, SVM, CRF, and Naïve Bayes, to predict the morphological types. On the average, the best algorithm CRF achieved 75% of the annotators’ performance.