With the recent advances of open-domain story generation, the lack of reliable automatic evaluation metrics becomes an increasingly imperative issue that hinders the fast development of story generation. According to conducted researches in this regard, learnable evaluation metrics have promised more accurate assessments by having higher correlations with human judgments. A critical bottleneck of obtaining a reliable learnable evaluation metric is the lack of high-quality training data for classifiers to efficiently distinguish plausible and implausible machine-generated stories. Previous works relied on \textit{heuristically manipulated} plausible examples to mimic possible system drawbacks such as repetition, contradiction, or irrelevant content in the text level, which can be \textit{unnatural} and \textit{oversimplify} the characteristics of implausible machine-generated stories. We propose to tackle these issues by generating a more comprehensive set of implausible stories using {\em plots}, which are structured representations of controllable factors used to generate stories. Since these plots are compact and structured, it is easier to manipulate them to generate text with targeted undesirable properties, while at the same time maintain the grammatical correctness and naturalness of the generated sentences. To improve the quality of generated implausible stories, we further apply the adversarial filtering procedure presented by \citet{zellers2018swag} to select a more nuanced set of implausible texts. Experiments show that the evaluation metrics trained on our generated data result in more reliable automatic assessments that correlate remarkably better with human judgments compared to the baselines.
Generating metaphors is a challenging task as it requires a proper understanding of abstract concepts, making connections between unrelated concepts, and deviating from the literal meaning. In this paper, we aim to generate a metaphoric sentence given a literal expression by replacing relevant verbs. Based on a theoretically-grounded connection between metaphors and symbols, we propose a method to automatically construct a parallel corpus by transforming a large number of metaphorical sentences from the Gutenberg Poetry corpus (Jacobs, 2018) to their literal counterpart using recent advances in masked language modeling coupled with commonsense inference. For the generation task, we incorporate a metaphor discriminator to guide the decoding of a sequence to sequence model fine-tuned on our parallel data to generate high-quality metaphors. Human evaluation on an independent test set of literal statements shows that our best model generates metaphors better than three well-crafted baselines 66% of the time on average. A task-based evaluation shows that human-written poems enhanced with metaphors proposed by our model are preferred 68% of the time compared to poems without metaphors.
Auto-regressive language models with the left-to-right generation order have been a predominant paradigm for language generation. Recently, out-of-order text generation beyond the traditional left-to-right paradigm has attracted extensive attention, with a notable variation of insertion-based generation, where a model is used to gradually extend the context into a complete sentence purely with insertion operations. However, since insertion operations disturb the position information of each token, it is often believed that each step of the insertion-based likelihood estimation requires a bi-directional \textit{re-encoding} of the whole generated sequence. This computational overhead prohibits the model from scaling up to generate long, diverse texts such as stories, news articles, and reports. To address this issue, we propose InsNet, an insertion-based sequence model that can be trained as efficiently as traditional transformer decoders while maintaining the same performance as that with a bi-directional context encoder. We evaluate InsNet on story generation and CleVR-CoGENT captioning, showing the advantages of InsNet in several dimensions, including computational costs, generation quality, the ability to perfectly incorporate lexical controls, and better compositional generalization.
Having engaging and informative conversations with users is the utmost goal for open-domain conversational systems. Recent advances in transformer-based language models and their applications to dialogue systems have succeeded to generate fluent and human-like responses. However, they still lack control over the generation process towards producing contentful responses and achieving engaging conversations. To achieve this goal, we present \textbf{DiSCoL} (\textbf{Di}alogue \textbf{S}ystems through \textbf{Co}versational \textbf{L}ine guided response generation). DiSCoL is an open-domain dialogue system that leverages conversational lines (briefly \textbf{convlines}) as controllable and informative content-planning elements to guide the generation model produce engaging and informative responses. Two primary modules in DiSCoL's pipeline are conditional generators trained for 1) predicting relevant and informative convlines for dialogue contexts and 2) generating high-quality responses conditioned on the predicted convlines. Users can also change the returned convlines to \textit{control} the direction of the conversations towards topics that are more interesting for them. Through automatic and human evaluations, we demonstrate the efficiency of the convlines in producing engaging conversations.
Even for domain experts, it is a non-trivial task to verify a scientific claim by providing supporting or refuting evidence rationales. The situation worsens as misinformation is proliferated on social media or news websites, manually or programmatically, at every moment. As a result, an automatic fact-verification tool becomes crucial for combating the spread of misinformation. In this work, we propose a novel, paragraph-level, multi-task learning model for the SciFact task by directly computing a sequence of contextualized sentence embeddings from a BERT model and jointly training the model on rationale selection and stance prediction.
We present EventPlus, a temporal event understanding pipeline that integrates various state-of-the-art event understanding components including event trigger and type detection, event argument detection, event duration and temporal relation extraction. Event information, especially event temporal knowledge, is a type of common sense knowledge that helps people understand how stories evolve and provides predictive hints for future events. EventPlus as the first comprehensive temporal event understanding pipeline provides a convenient tool for users to quickly obtain annotations about events and their temporal information for any user-provided document. Furthermore, we show EventPlus can be easily adapted to other domains (e.g., biomedical domain). We make EventPlus publicly available to facilitate event-related information extraction and downstream applications.
The ability to capture complex linguistic structures and long-term dependencies among words in the passage is essential for many natural language understanding tasks. In relation extraction, dependency trees that contain rich syntactic clues have been widely used to help capture long-term dependencies in text. Graph neural networks (GNNs), one of the means to encode dependency graphs, has been shown effective in several prior works. However, relatively little attention has been paid to the receptive fields of GNNs, which can be crucial in tasks with extremely long text that go beyond single sentences and require discourse analysis. In this work, we leverage the idea of graph pooling and propose the Mirror Graph Convolution Network (MrGCN), a GNN model with pooling-unpooling structures tailored to relation extraction. The pooling branch reduces the graph size and enables the GCN to obtain larger receptive fields within less layers; the unpooling branch restores the pooled graph to its original resolution such that token-level relation extraction can be performed. Experiments on two datasets demonstrate the effectiveness of our method, showing significant improvements over previous results.
Pretrained language models (LMs) such as BERT, RoBERTa, and ELECTRA are effective at improving the performances of a variety of downstream NLP tasks. Recently, researchers have incorporated domain and task-specific knowledge in these LMs' training objectives and further enhanced models' capability of handling downstream tasks. However, none of these LMs are designed specifically for event temporal reasoning. We propose DEER, a language model that is trained to focus on event temporal relations and performs better under low-resource settings than original LMs. More specifically, we create a large number of training samples to simulate the machine reading comprehension and information extraction tasks for event temporal understanding and leverage a generator-discriminator structure to reinforce the LMs' capability of event temporal reasoning. Our experimental results show that DEER can achieve SOTA results and works particularly well in low-resource settings across 5 widely used datasets.