Open domain question answering systems frequently rely on information retrieved from large collections of text (such as the Web) to answer questions. However, such collections of text often contain conflicting information, and indiscriminately depending on this information may result in untruthful and inaccurate answers. To understand the gravity of this problem, we collect a human-annotated dataset, Question Answering with Conflicting Contexts (QACC), and find that as much as 25% of unambiguous, open domain questions can lead to conflicting contexts when retrieved using Google Search. We evaluate and benchmark three powerful Large Language Models (LLMs) with our dataset QACC and demonstrate their limitations in effectively addressing questions with conflicting information. To explore how humans reason through conflicting contexts, we request our annotators to provide explanations for their selections of correct answers. We demonstrate that by finetuning LLMs to explain their answers, we can introduce richer information into their training that guide them through the process of reasoning with conflicting contexts. We publicly release our dataset and code to promote research along this line.

Large Language Models (LLMs) have demonstrated remarkable performance across various tasks. However, they are prone to contextual hallucination, generating information that is either unsubstantiated or contradictory to the given context. Although many studies have investigated contextual hallucinations in LLMs, addressing them in long-context inputs remains an open problem. In this work, we take an initial step toward solving this problem by constructing a dataset specifically designed for long-context hallucination detection. Furthermore, we propose a novel architecture that enables pre-trained encoder models, such as BERT, to process long contexts and effectively detect contextual hallucinations through a decomposition and aggregation mechanism. Our experimental results show that the proposed architecture significantly outperforms previous models of similar size as well as LLM-based models across various metrics, while providing substantially faster inference. We publicly release our dataset and code to promote research along the same line.

Missing information is a common issue of dialogue summarization where some information in the reference summaries is not covered in the generated summaries. To address this issue, we propose to utilize natural language inference (NLI) models to improve coverage while avoiding introducing factual inconsistencies. Specifically, we use NLI to compute fine-grained training signals to encourage the model to generate content in the reference summaries that have not been covered, as well as to distinguish between factually consistent and inconsistent generated sentences. Experiments on the DialogSum and SAMSum datasets confirm the effectiveness of the proposed approach in balancing coverage and faithfulness, validated with automatic metrics and human evaluations. Additionally, we compute the correlation between commonly used automatic metrics with human judgments in terms of three different dimensions regarding coverage and factual consistency to provide insight into the most suitable metric for evaluating dialogue summaries.

Pretrained language models (PTLMs) are typically learned over a large, static corpus and further fine-tuned for various downstream tasks. However, when deployed in the real world, a PTLM-based model must deal with data distributions that deviates from what the PTLM was initially trained on. In this paper, we study a lifelong language model pretraining challenge where a PTLM is continually updated so as to adapt to emerging data. Over a domain-incremental research paper stream and a chronologically-ordered tweet stream, we incrementally pretrain a PTLM with different continual learning algorithms, and keep track of the downstream task performance (after fine-tuning). We evaluate PTLM’s ability to adapt to new corpora while retaining learned knowledge in earlier corpora. Our experiments show distillation-based approaches to be most effective in retaining downstream performance in earlier domains. The algorithms also improve knowledge transfer, allowing models to achieve better downstream performance over latest data, and improve temporal generalization when distribution gaps exist between training and evaluation because of time. We believe our problem formulation, methods, and analysis will inspire future studies towards continual pretraining of language models.

Despite profound successes, contrastive representation learning relies on carefully designed data augmentations using domain-specific knowledge. This challenge is magnified in natural language processing, where no general rules exist for data augmentation due to the discrete nature of natural language. We tackle this challenge by presenting a Virtual augmentation Supported Contrastive Learning of sentence representations (VaSCL). Originating from the interpretation that data augmentation essentially constructs the neighborhoods of each training instance, we, in turn, utilize the neighborhood to generate effective data augmentations. Leveraging the large training batch size of contrastive learning, we approximate the neighborhood of an instance via its K-nearest in-batch neighbors in the representation space. We then define an instance discrimination task regarding the neighborhood and generate the virtual augmentation in an adversarial training manner. We access the performance of VaSCL on a wide range of downstream tasks and set a new state-of-the-art for unsupervised sentence representation learning.

Learning high-quality dialogue representations is essential for solving a variety of dialogue-oriented tasks, especially considering that dialogue systems often suffer from data scarcity. In this paper, we introduce Dialogue Sentence Embedding (DSE), a self-supervised contrastive learning method that learns effective dialogue representations suitable for a wide range of dialogue tasks. DSE learns from dialogues by taking consecutive utterances of the same dialogue as positive pairs for contrastive learning. Despite its simplicity, DSE achieves significantly better representation capability than other dialogue representation and universal sentence representation models. We evaluate DSE on five downstream dialogue tasks that examine dialogue representation at different semantic granularities. Experiments in few-shot and zero-shot settings show that DSE outperforms baselines by a large margin, for example, it achieves 13% average performance improvement over the strongest unsupervised baseline in 1-shot intent classification on 6 datasets. We also provide analyses on the benefits and limitations of our model.

Pretrained language models (PTLMs) are typically learned over a large, static corpus and further fine-tuned for various downstream tasks. However, when deployed in the real world, a PTLM-based model must deal with data distributions that deviates from what the PTLM was initially trained on. In this paper, we study a lifelong language model pretraining challenge where a PTLM is continually updated so as to adapt to emerging data. Over a domain-incremental research paper stream and a chronologically-ordered tweet stream, we incrementally pretrain a PTLM with different continual learning algorithms, and keep track of the downstream task performance (after fine-tuning). We evaluate PTLM’s ability to adapt to new corpora while retaining learned knowledge in earlier corpora. Our experiments show distillation-based approaches to be most effective in retaining downstream performance in earlier domains. The algorithms also improve knowledge transfer, allowing models to achieve better downstream performance over latest data, and improve temporal generalization when distribution gaps exist between training and evaluation because of time. We believe our problem formulation, methods, and analysis will inspire future studies towards continual pretraining of language models.

Many recent successes in sentence representation learning have been achieved by simply fine-tuning on the Natural Language Inference (NLI) datasets with triplet loss or siamese loss. Nevertheless, they share a common weakness: sentences in a contradiction pair are not necessarily from different semantic categories. Therefore, optimizing the semantic entailment and contradiction reasoning objective alone is inadequate to capture the high-level semantic structure. The drawback is compounded by the fact that the vanilla siamese or triplet losses only learn from individual sentence pairs or triplets, which often suffer from bad local optima. In this paper, we propose PairSupCon, an instance discrimination based approach aiming to bridge semantic entailment and contradiction understanding with high-level categorical concept encoding. We evaluate PairSupCon on various downstream tasks that involve understanding sentence semantics at different granularities. We outperform the previous state-of-the-art method with 10%–13% averaged improvement on eight clustering tasks, and 5%–6% averaged improvement on seven semantic textual similarity (STS) tasks.

Intelligent personal assistants (IPAs) such as Amazon Alexa, Google Assistant and Apple Siri extend their built-in capabilities by supporting voice apps developed by third-party developers. Sometimes the smart assistant is not able to successfully respond to user voice commands (aka utterances). There are many reasons including automatic speech recognition (ASR) error, natural language understanding (NLU) error, routing utterances to an irrelevant voice app or simply that the user is asking for a capability that is not supported yet. The failure to handle a voice command leads to customer frustration. In this paper, we introduce a fallback skill recommendation system to suggest a voice app to a customer for an unhandled voice command. One of the prominent challenges of developing a skill recommender system for IPAs is partial observation. To solve the partial observation problem, we propose collaborative data relabeling (CDR) method. In addition, CDR also improves the diversity of the recommended skills. We evaluate the proposed method both offline and online. The offline evaluation results show that the proposed system outperforms the baselines. The online A/B testing results show significant gain of customer experience metrics.

The ability to automatically and accurately process customer feedback is a necessity in the private sector. Unfortunately, customer feedback can be one of the most difficult types of data to work with due to the sheer volume and variety of services, products, languages, and cultures that comprise the customer experience. In order to address this issue, our team built a suite of classifiers trained on a four-language, multi-label corpus released as part of the shared task on “Customer Feedback Analysis” at IJCNLP 2017. In addition to standard text preprocessing, we translated each dataset into each other language to increase the size of the training datasets. Additionally, we also used word embeddings in our feature engineering step. Ultimately, we trained classifiers using Logistic Regression, Random Forest, and Long Short-Term Memory (LSTM) Recurrent Neural Networks. Overall, we achieved a Macro-Average F-score between 48.7% and 56.0% for the four languages and ranked 3/12 for English, 3/7 for Spanish, 1/8 for French, and 2/7 for Japanese.