Artificial Intelligence (AI), along with the recent progress in biomedical language understanding, is gradually changing medical practice. With the development of biomedical language understanding benchmarks, AI applications are widely used in the medical field. However, most benchmarks are limited to English, which makes it challenging to replicate many of the successes in English for other languages. To facilitate research in this direction, we collect real-world biomedical data and present the first Chinese Biomedical Language Understanding Evaluation (CBLUE) benchmark: a collection of natural language understanding tasks including named entity recognition, information extraction, clinical diagnosis normalization, single-sentence/sentence-pair classification, and an associated online platform for model evaluation, comparison, and analysis. To establish evaluation on these tasks, we report empirical results with the current 11 pre-trained Chinese models, and experimental results show that state-of-the-art neural models perform by far worse than the human ceiling. Our benchmark is released at \url{https://tianchi.aliyun.com/dataset/dataDetail?dataId=95414&lang=en-us}.
Data augmentation with mixup has shown to be effective on various computer vision tasks. Despite its great success, there has been a hurdle to apply mixup to NLP tasks since text consists of discrete tokens with variable length. In this work, we propose SSMix, a novel mixup method where the operation is performed on input text rather than on hidden vectors like previous approaches. SSMix synthesizes a sentence while preserving the locality of two original texts by span-based mixing and keeping more tokens related to the prediction relying on saliency information. With extensive experiments, we empirically validate that our method outperforms hidden-level mixup methods on a wide range of text classification benchmarks, including textual entailment, sentiment classification, and question-type classification. Our code is available at https://github.com/clovaai/ssmix.
We exhibit that the implicit UCCA parser does not address numeric fused-heads (NFHs) consistently, which could result either from inconsistent annotation, insufficient training data or a modelling limitation. and show which factors are involved. We consider this phenomenon important, as it is pervasive in text and critical for correct inference. Careful design and fine-grained annotation of NFHs in meaning representation frameworks would benefit downstream tasks such as machine translation, natural language inference and question answering, particularly when they require numeric reasoning, as recovering and categorizing them. We are investigating the treatment of this phenomenon by other meaning representations, such as AMR. We encourage researchers in meaning representations, and computational linguistics in general, to address this phenomenon in future research.
Pre-trained language models such as ClinicalBERT have achieved impressive results on tasks such as medical Natural Language Inference. At first glance, this may suggest that these models are able to perform medical reasoning tasks, such as mapping symptoms to diseases. However, we find that standard benchmarks such as MedNLI contain relatively few examples that require such forms of reasoning. To better understand the medical reasoning capabilities of existing language models, in this paper we introduce DisKnE, a new benchmark for Disease Knowledge Evaluation. To construct this benchmark, we annotated each positive MedNLI example with the types of medical reasoning that are needed. We then created negative examples by corrupting these positive examples in an adversarial way. Furthermore, we define training-test splits per disease, ensuring that no knowledge about test diseases can be learned from the training data, and we canonicalize the formulation of the hypotheses to avoid the presence of artefacts. This leads to a number of binary classification problems, one for each type of reasoning and each disease. When analysing pre-trained models for the clinical/biomedical domain on the proposed benchmark, we find that their performance drops considerably.
Recursive Neural Networks (RvNNs), which compose sequences according to their underlying hierarchical syntactic structure, have performed well in several natural language processing tasks compared to similar models without structural biases. However, traditional RvNNs are incapable of inducing the latent structure in a plain text sequence on their own. Several extensions have been proposed to overcome this limitation. Nevertheless, these extensions tend to rely on surrogate gradients or reinforcement learning at the cost of higher bias or variance. In this work, we propose Continuous Recursive Neural Network (CRvNN) as a backpropagation-friendly alternative to address the aforementioned limitations. This is done by incorporating a continuous relaxation to the induced structure. We demonstrate that CRvNN achieves strong performance in challenging synthetic tasks such as logical inference and ListOps. We also show that CRvNN performs comparably or better than prior latent structure models on real-world tasks such as sentiment analysis and natural language inference.
While large-scale pretrained language models have obtained impressive results when fine-tuned on a wide variety of tasks, they still often suffer from overfitting in low-resource scenarios. Since such models are general-purpose feature extractors, many of these features are inevitably irrelevant for a given target task. We propose to use Variational Information Bottleneck (VIB) to suppress irrelevant features when fine-tuning on low-resource target tasks, and show that our method successfully reduces overfitting. Moreover, we show that our VIB model finds sentence representations that are more robust to biases in natural language inference datasets, and thereby obtains better generalization to out-of-domain datasets. Evaluation on seven low-resource datasets in different tasks shows that our method significantly improves transfer learning in low-resource scenarios, surpassing prior work. Moreover, it improves generalization on 13 out of 15 out-of-domain natural language inference benchmarks. Our code is publicly available in https://github.com/rabeehk/vibert.
Multilingual transformers (XLM, mT5) have been shown to have remarkable transfer skills in zero-shot settings. Most transfer studies, however, rely on automatically translated resources (XNLI, XQuAD), making it hard to discern the particular linguistic knowledge that is being transferred, and the role of expert annotated monolingual datasets when developing task-specific models. We investigate the cross-lingual transfer abilities of XLM-R for Chinese and English natural language inference (NLI), with a focus on the recent large-scale Chinese dataset OCNLI. To better understand linguistic transfer, we created 4 categories of challenge and adversarial tasks (totaling 17 new datasets) for Chinese that build on several well-known resources for English (e.g., HANS, NLI stress-tests). We find that cross-lingual models trained on English NLI do transfer well across our Chinese tasks (e.g., in 3/4 of our challenge categories, they perform as well/better than the best monolingual models, even on 3/5 uniquely Chinese linguistic phenomena such as idioms, pro drop). These results, however, come with important caveats: cross-lingual models often perform best when trained on a mixture of English and high-quality monolingual NLI data (OCNLI), and are often hindered by automatically translated resources (XNLI-zh). For many phenomena, all models continue to struggle, highlighting the need for our new diagnostics to help benchmark Chinese and cross-lingual models. All new datasets/code are released at https://github.com/huhailinguist/ChineseNLIProbing.
In this report, we introduce SciFive, a domain-specific T5 model that has been pre-trained on large biomedical corpora. Our model outperforms the current SOTA methods (i.e. BERT, BioBERT, Base T5) on tasks in named entity relation, relation extraction, natural language inference, and question-answering. We show that text-generation methods have significant potential in a broad array of biomedical NLP tasks, particularly those requiring longer, more complex outputs. Our results support the exploration of more difficult text generation tasks and the development of new methods in this area
Natural Language Inference (NLI) datasets contain examples with highly ambiguous labels. While many research works do not pay much attention to this fact, several recent efforts have been made to acknowledge and embrace the existence of ambiguity, such as UNLI and ChaosNLI. In this paper, we explore the option of training directly on the estimated label distribution of the annotators in the NLI task, using a learning loss based on this ambiguity distribution instead of the gold-labels. We prepare AmbiNLI, a trial dataset obtained from readily available sources, and show it is possible to reduce ChaosNLI divergence scores when finetuning on this data, a promising first step towards learning how to capture linguistic ambiguity. Additionally, we show that training on the same amount of data but targeting the ambiguity distribution instead of gold-labels can result in models that achieve higher performance and learn better representations for downstream tasks.
Although deep learning models have driven state-of-the-art performance on a wide array of tasks, they are prone to spurious correlations that should not be learned as predictive clues. To mitigate this problem, we propose a causality-based training framework to reduce the spurious correlations caused by observed confounders. We give theoretical analysis on the underlying general Structural Causal Model (SCM) and propose to perform Maximum Likelihood Estimation (MLE) on the interventional distribution instead of the observational distribution, namely Counterfactual Maximum Likelihood Estimation (CMLE). As the interventional distribution, in general, is hidden from the observational data, we then derive two different upper bounds of the expected negative log-likelihood and propose two general algorithms, Implicit CMLE and Explicit CMLE, for causal predictions of deep learning models using observational data. We conduct experiments on both simulated data and two real-world tasks: Natural Language Inference (NLI) and Image Captioning. The results show that CMLE methods outperform the regular MLE method in terms of out-of-domain generalization performance and reducing spurious correlations, while maintaining comparable performance on the regular evaluations.
Present language understanding methods have demonstrated extraordinary ability of recognizing patterns in texts via machine learning. However, existing methods indiscriminately use the recognized patterns in the testing phase that is inherently different from us humans who have counterfactual thinking, e.g., to scrutinize for the hard testing samples. Inspired by this, we propose a Counterfactual Reasoning Model, which mimics the counterfactual thinking by learning from few counterfactual samples. In particular, we devise a generation module to generate representative counterfactual samples for each factual sample, and a retrospective module to retrospect the model prediction by comparing the counterfactual and factual samples. Extensive experiments on sentiment analysis (SA) and natural language inference (NLI) validate the effectiveness of our method.
Structural analysis methods (e.g., probing and feature attribution) are increasingly important tools for neural network analysis. We propose a new structural analysis method grounded in a formal theory of causal abstraction that provides rich characterizations of model-internal representations and their roles in input/output behavior. In this method, neural representations are aligned with variables in interpretable causal models, and then interchange interventions are used to experimentally verify that the neural representations have the causal properties of their aligned variables. We apply this method in a case study to analyze neural models trained on Multiply Quantified Natural Language Inference (MQNLI) corpus, a highly complex NLI dataset that was constructed with a tree-structured natural logic causal model. We discover that a BERT-based model with state-of-the-art performance successfully realizes parts of the natural logic model's causal structure, whereas a simpler baseline model fails to show any such structure, demonstrating that BERT representations encode the compositional structure of MQNLI.
Cross-lingual language tasks typically require a substantial amount of annotated data or parallel translation data. We explore whether language representations that capture relationships among languages can be learned and subsequently leveraged in cross-lingual tasks without the use of parallel data. We generate dense embeddings for 29 languages using a denoising autoencoder, and evaluate the embeddings using the World Atlas of Language Structures (WALS) and two extrinsic tasks in a zero-shot setting: cross-lingual dependency parsing and cross-lingual natural language inference.
A good open-domain chatbot should avoid presenting contradictory responses about facts or opinions in a conversational session, known as its consistency capacity. However, evaluating the consistency capacity of a chatbot is still challenging. Employing human judges to interact with chatbots on purpose to check their capacities is costly and low-efficient, and difficult to get rid of subjective bias. In this paper, we propose the Addressing Inquiries about History (AIH), an efficient and practical framework for the consistency evaluation. At the conversation stage, AIH attempts to address appropriate inquiries about the dialogue history to induce the chatbot to redeclare the historical facts or opinions. We carry out the conversation between chatbots, which is more efficient than the human-bot interaction and can also alleviate the subjective bias. In this way, we manage to rapidly obtain a dialog session that contains responses with high contradiction possibilities. At the contradiction recognition stage, we can either employ human judges or a natural language inference (NLI) model to recognize whether the answers to the inquiries are contradictory with history. Finally, we are able to rank chatbots according to the contradiction statistics. Experiments on open-domain chatbots show that our approach can efficiently and reliably assess the consistency capacity of chatbots and achieve a high ranking correlation with the human evaluation. We release the framework and hope to help improve the consistency capacity of chatbots. \footnote{\url{https://github.com/ictnlp/AIH}}
The ability to learn from limited data, or few-shot learning, is a desirable and often critical requirement for NLP systems. While many existing methods do poorly at learning from a handful of examples, large pretrained language models have recently been shown to be efficient few-shot learners. One approach to few-shot learning, which does not require finetuning of model parameters, is to augment the language model's input with priming text which is typically constructed using task specific descriptions and examples. In this work, we further explore priming-based few-shot learning, with focus on using examples as prompts. We show that presenting examples in the right order is key for generalization. We introduce PERO (Prompting with Examples in the Right Order), where we formulate few-shot learning as search over the set of permutations of the training examples. We show that PERO can learn to generalize efficiently using as few as 10 examples, in contrast to existing approaches. While the newline token is a natural choice for separating the examples in the prompt, we show that learning a new separator token can potentially provide further gains in performance. We demonstrate the effectiveness of the proposed method on the tasks of sentiment classification, natural language inference and fact retrieval. Finally, we analyze the learned prompts to reveal novel insights, including the idea that two training examples in the right order alone can provide competitive performance for sentiment classification and natural language inference.
Crowdworker-constructed natural language inference (NLI) datasets have been found to contain statistical artifacts associated with the annotation process that allow hypothesis-only classifiers to achieve better-than-random performance (Poliak et al., 2018; Gururanganet et al., 2018; Tsuchiya, 2018). We investigate whether MedNLI, a physician-annotated dataset with premises extracted from clinical notes, contains such artifacts (Romanov and Shivade, 2018). We find that entailed hypotheses contain generic versions of specific concepts in the premise, as well as modifiers related to responsiveness, duration, and probability. Neutral hypotheses feature conditions and behaviors that co-occur with, or cause, the condition(s) in the premise. Contradiction hypotheses feature explicit negation of the premise and implicit negation via assertion of good health. Adversarial filtering demonstrates that performance degrades when evaluated on the difficult subset. We provide partition information and recommendations for alternative dataset construction strategies for knowledge-intensive domains.
We introduce a collection of recognizing textual entailment (RTE) datasets focused on figurative language. We leverage five existing datasets annotated for a variety of figurative language -- simile, metaphor, and irony -- and frame them into over 12,500 RTE examples.We evaluate how well state-of-the-art models trained on popular RTE datasets capture different aspects of figurative language. Our results and analyses indicate that these models might not sufficiently capture figurative language, struggling to perform pragmatic inference and reasoning about world knowledge. Ultimately, our datasets provide a challenging testbed for evaluating RTE models.
Commonsense inference to understand and explain human language is a fundamental research problem in natural language processing. Explaining human conversations poses a great challenge as it requires contextual understanding, planning, inference, and several aspects of reasoning including causal, temporal, and commonsense reasoning. In this work, we introduce CIDER -- a manually curated dataset that contains dyadic dialogue explanations in the form of implicit and explicit knowledge triplets inferred using contextual commonsense inference. Extracting such rich explanations from conversations can be conducive to improving several downstream applications. The annotated triplets are categorized by the type of commonsense knowledge present (e.g., causal, conditional, temporal). We set up three different tasks conditioned on the annotated dataset: Dialogue-level Natural Language Inference, Span Extraction, and Multi-choice Span Selection. Baseline results obtained with transformer-based models reveal that the tasks are difficult, paving the way for promising future research. The dataset and the baseline implementations are publicly available at https://cider-task.github.io/cider/.
Deep learning (DL) based language models achieve high performance on various benchmarks for Natural Language Inference (NLI). And at this time, symbolic approaches to NLI are receiving less attention. Both approaches (symbolic and DL) have their advantages and weaknesses. However, currently, no method combines them in a system to solve the task of NLI. To merge symbolic and deep learning methods, we propose an inference framework called NeuralLog, which utilizes both a monotonicity-based logical inference engine and a neural network language model for phrase alignment. Our framework models the NLI task as a classic search problem and uses the beam search algorithm to search for optimal inference paths. Experiments show that our joint logic and neural inference system improves accuracy on the NLI task and can achieve state-of-art accuracy on the SICK and MED datasets.
Automated event extraction in social science applications often requires corpus-level evaluations: for example, aggregating text predictions across metadata and unbiased estimates of recall. We combine corpus-level evaluation requirements with a real-world, social science setting and introduce the IndiaPoliceEvents corpus--all 21,391 sentences from 1,257 English-language Times of India articles about events in the state of Gujarat during March 2002. Our trained annotators read and label every document for mentions of police activity events, allowing for unbiased recall evaluations. In contrast to other datasets with structured event representations, we gather annotations by posing natural questions, and evaluate off-the-shelf models for three different tasks: sentence classification, document ranking, and temporal aggregation of target events. We present baseline results from zero-shot BERT-based models fine-tuned on natural language inference and passage retrieval tasks. Our novel corpus-level evaluations and annotation approach can guide creation of similar social-science-oriented resources in the future.