Human communication relies on common ground (CG), the mutual knowledge and beliefs shared by participants, to produce coherent and interesting conversations. In this paper, we demonstrate that current response generation (RG) models produce generic and dull responses in dialogues because they act reflexively, failing to explicitly model CG, both due to the lack of CG in training data and the standard RG training procedure. We introduce Reflect, a dataset that annotates dialogues with explicit CG (materialized as inferences approximating shared knowledge and beliefs) and solicits 9k diverse human-generated responses each following one common ground. Using Reflect, we showcase the limitations of current dialogue data and RG models: less than half of the responses in current data are rated as high quality (sensible, specific, and interesting) and models trained using this data have even lower quality, while most Reflect responses are judged high quality. Next, we analyze whether CG can help models produce better-quality responses by using Reflect CG to guide RG models. Surprisingly, we find that simply prompting GPT3 to "think" about CG generates 30% more quality responses, showing promising benefits to integrating CG into the RG process.
Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 442 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit "breakthrough" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting.
Real-world natural language processing (NLP) models need to be continually updated to fix the prediction errors in out-of-distribution (OOD) data streams while overcoming catastrophic forgetting. However, existing continual learning (CL) problem setups cannot cover such a realistic and complex scenario. In response to this, we propose a new CL problem formulation dubbed continual model refinement (CMR). Compared to prior CL settings, CMR is more practical and introduces unique challenges (boundary-agnostic and non-stationary distribution shift, diverse mixtures of multiple OOD data clusters, error-centric streams, etc.). We extend several existing CL approaches to the CMR setting and evaluate them extensively. For benchmarking and analysis, we propose a general sampling algorithm to obtain dynamic OOD data streams with controllable non-stationarity, as well as a suite of metrics measuring various aspects of online performance. Our experiments and detailed analysis reveal the promise and challenges of the CMR problem, supporting that studying CMR in dynamic OOD streams can benefit the longevity of deployed NLP models in production.
Humans can perform unseen tasks by recalling relevant skills that are acquired previously and then generalizing them to the target tasks, even if there is no supervision at all. In this paper, we aim to improve such cross-task generalization ability of massive multi-task language models such as T0 (Sanh et al., 2021) in an unsupervised setting. We propose a retrieval-augmentation method named ReCross that takes a few unlabelled examples as queries to retrieve a small subset of upstream data and uses them to update the multi-task model for better generalization. Our empirical results show that the proposed ReCross consistently outperforms non-retrieval baselines by a significant margin.
We study the robustness of machine reading comprehension (MRC) models to entity renaming -- do models make more wrong predictions when answer entities have different names? Such failures would indicate that models are overly reliant on entity knowledge to answer questions, and therefore may generalize poorly when facts about the world change or questions are asked about novel entities. To systematically audit model robustness, we propose a general and scalable method to replace person names with names from a variety of sources, ranging from common English names to names from other languages to arbitrary strings. Across four datasets and three pretrained model architectures, MRC models consistently perform worse when entities are renamed, with particularly large accuracy drops on datasets constructed via distant supervision. We also find large differences between models: SpanBERT, which is pretrained with span-level masking, is more robust than RoBERTa, despite having similar accuracy on unperturbed test data. Inspired by this, we experiment with span-level and entity-level masking as a continual pretraining objective and find that they can further improve the robustness of MRC models.
To audit the robustness of named entity recognition (NER) models, we propose RockNER, a simple yet effective method to create natural adversarial examples. Specifically, at the entity level, we replace target entities with other entities of the same semantic class in Wikidata; at the context level, we use pre-trained language models (e.g., BERT) to generate word substitutions. Together, the two levels of attack produce natural adversarial examples that result in a shifted distribution from the training data on which our target models have been trained. We apply the proposed method to the OntoNotes dataset and create a new benchmark named OntoRock for evaluating the robustness of existing NER models via a systematic evaluation protocol. Our experiments and analysis reveal that even the best model has a significant performance drop, and these models seem to memorize in-domain entity patterns instead of reasoning from the context. Our work also studies the effects of a few simple data augmentation methods to improve the robustness of NER models.
Deep neural models for low-resource named entity recognition (NER) have shown impressive results by leveraging distant super-vision or other meta-level information (e.g. explanation). However, the costs of acquiring such additional information are generally prohibitive, especially in domains where existing resources (e.g. databases to be used for distant supervision) may not exist. In this paper, we present a novel two-stage framework (AutoTriggER) to improve NER performance by automatically generating and leveraging "entity triggers" which are essentially human-readable clues in the text that can help guide the model to make better decisions. Thus, the framework is able to both create and leverage auxiliary supervision by itself. Through experiments on three well-studied NER datasets, we show that our automatically extracted triggers are well-matched to human triggers, and AutoTriggER improves performance over a RoBERTa-CRFarchitecture by nearly 0.5 F1 points on average and much more in a low resource setting.
Commonsense reasoning research has so far been limited to English. We aim to evaluate and improve popular multilingual language models (ML-LMs) to help advance commonsense reasoning (CSR) beyond English. We collect the Mickey Corpus, consisting of 561k sentences in 11 different languages, which can be used for analyzing and improving ML-LMs. We propose Mickey Probe, a language-agnostic probing task for fairly evaluating the common sense of popular ML-LMs across different languages. In addition, we also create two new datasets, X-CSQA and X-CODAH, by translating their English versions to 15 other languages, so that we can evaluate popular ML-LMs for cross-lingual commonsense reasoning. To improve the performance beyond English, we propose a simple yet effective method -- multilingual contrastive pre-training (MCP). It significantly enhances sentence representations, yielding a large performance gain on both benchmarks.
Communication is a cooperative effort that requires reaching mutual understanding among the participants. Humans use commonsense reasoning implicitly to produce natural and logically-coherent responses. As a step towards fluid human-AI communication, we study if response generation (RG) models can emulate human reasoning process and use common sense to help produce better-quality responses. We aim to tackle two research questions: how to formalize conversational common sense and how to examine RG models capability to use common sense? We first propose a task, CEDAR: Causal common sEnse in DiAlogue Response generation, that concretizes common sense as textual explanations for what might lead to the response and evaluates RG models behavior by comparing the modeling loss given a valid explanation with an invalid one. Then we introduce a process that automatically generates such explanations and ask humans to verify them. Finally, we design two probing settings for RG models targeting two reasoning capabilities using verified explanations. We find that RG models have a hard time determining the logical validity of explanations but can identify grammatical naturalness of the explanation easily.