Recently, aspect sentiment quad prediction (ASQP) has become a popular task in the field of aspect-level sentiment analysis. Previous work utilizes a predefined template to paraphrase the original sentence into a structure target sequence, which can be easily decoded as quadruplets of the form (aspect category, aspect term, opinion term, sentiment polarity). The template involves the four elements in a fixed order. However, we observe that this solution contradicts with the order-free property of the ASQP task, since there is no need to fix the template order as long as the quadruplet is extracted correctly. Inspired by the observation, we study the effects of template orders and find that some orders help the generative model achieve better performance. It is hypothesized that different orders provide various views of the quadruplet. Therefore, we propose a simple but effective method to identify the most proper orders, and further combine multiple proper templates as data augmentation to improve the ASQP task. Specifically, we use the pre-trained language model to select the orders with minimal entropy. By fine-tuning the pre-trained language model with these template orders, our approach improves the performance of quad prediction, and outperforms state-of-the-art methods significantly in low-resource settings.
Recently introduced language model prompting methods can achieve high accuracy in zero- and few-shot settings while requiring few to no learned task-specific parameters. Nevertheless, these methods still often trail behind full model finetuning. In this work, we investigate if a dedicated continued pretraining stage could improve "promptability", i.e., zero-shot performance with natural language prompts or few-shot performance with prompt tuning. We reveal settings where existing continued pretraining methods lack promptability. We also identify current methodological gaps, which we fill with thorough large-scale experiments. We demonstrate that a simple recipe, continued pretraining that incorporates a trainable prompt during multi-task learning, leads to improved promptability in both zero- and few-shot settings compared to existing methods, up to 31% relative. On the other hand, we find that continued pretraining using MAML-style meta-learning, a method that directly optimizes few-shot promptability, yields subpar performance. We validate our findings with two prompt tuning methods, and, based on our results, we provide concrete recommendations to optimize promptability for different use cases.
In this paper, we present our submission to the sentence-level MQM benchmark at Quality Estimation Shared Task, named UniTE (Unified Translation Evaluation). Specifically, our systems employ the framework of UniTE, which combined three types of input formats during training with a pre-trained language model. First, we apply the pseudo-labeled data examples for the continuously pre-training phase. Notably, to reduce the gap between pre-training and fine-tuning, we use data pruning and a ranking-based score normalization strategy. For the fine-tuning phase, we use both Direct Assessment (DA) and Multidimensional Quality Metrics (MQM) data from past years' WMT competitions. Finally, we collect the source-only evaluation results, and ensemble the predictions generated by two UniTE models, whose backbones are XLM-R and InfoXLM, respectively. Results show that our models reach 1st overall ranking in the Multilingual and English-Russian settings, and 2nd overall ranking in English-German and Chinese-English settings, showing relatively strong performances in this year's quality estimation competition.
We study the application of large language models to zero-shot and few-shot classification of tabular data. We prompt the large language model with a serialization of the tabular data to a natural-language string, together with a short description of the classification problem. In the few-shot setting, we fine-tune the large language model using some labeled examples. We evaluate several serialization methods including templates, table-to-text models, and large language models. Despite its simplicity, we find that this technique outperforms prior deep-learning-based tabular classification methods on several benchmark datasets. In most cases, even zero-shot classification obtains non-trivial performance, illustrating the method's ability to exploit prior knowledge encoded in large language models. Unlike many deep learning methods for tabular datasets, this approach is also competitive with strong traditional baselines like gradient-boosted trees, especially in the very-few-shot setting.
Transformer-based Language Models (LMs) have achieved impressive results on natural language understanding tasks, but they can also generate toxic text such as insults, threats, and profanity, limiting their real-world applications. To overcome this issue, a few text generation approaches aim to detoxify toxic texts using additional LMs or perturbations. However, previous methods require excessive memory, computations, and time which are serious bottlenecks in their real-world application. To address such limitations, we propose an effective yet efficient method for language detoxification using an attribute-discriminative latent space. Specifically, we project the latent space of an original Transformer LM onto a discriminative latent space that well-separates texts by their attributes using a projection block and an attribute discriminator. This allows the LM to control the text generation to be non-toxic with minimal memory and computation overhead. We validate our model, Attribute-Discriminative Language Model (ADLM) on detoxified language and dialogue generation tasks, on which our method significantly outperforms baselines both in performance and efficiency.
Pre-trained language models (LMs), such as BERT (Devlin et al., 2018) and its variants, have led to significant improvements on various NLP tasks in past years. However, a theoretical framework for studying their relationships is still missing. In this paper, we fill this gap by investigating the linear dependency between pre-trained LMs. The linear dependency of LMs is defined analogously to the linear dependency of vectors. We propose Language Model Decomposition (LMD) to represent a LM using a linear combination of other LMs as basis, and derive the closed-form solution. A goodness-of-fit metric for LMD similar to the coefficient of determination is defined and used to measure the linear dependency of a set of LMs. In experiments, we find that BERT and eleven (11) BERT-like LMs are 91% linearly dependent. This observation suggests that current state-of-the-art (SOTA) LMs are highly "correlated". To further advance SOTA we need more diverse and novel LMs that are less dependent on existing LMs.
Most existing pre-trained language representation models (PLMs) are sub-optimal in sentiment analysis tasks, as they capture the sentiment information from word-level while under-considering sentence-level information. In this paper, we propose SentiWSP, a novel Sentiment-aware pre-trained language model with combined Word-level and Sentence-level Pre-training tasks. The word level pre-training task detects replaced sentiment words, via a generator-discriminator framework, to enhance the PLM's knowledge about sentiment words. The sentence level pre-training task further strengthens the discriminator via a contrastive learning framework, with similar sentences as negative samples, to encode sentiments in a sentence. Extensive experimental results show that SentiWSP achieves new state-of-the-art performance on various sentence-level and aspect-level sentiment classification benchmarks. We have made our code and model publicly available at https://github.com/XMUDM/SentiWSP.
In this report, we present our submission to the WMT 2022 Metrics Shared Task. We build our system based on the core idea of UNITE (Unified Translation Evaluation), which unifies source-only, reference-only, and source-reference-combined evaluation scenarios into one single model. Specifically, during the model pre-training phase, we first apply the pseudo-labeled data examples to continuously pre-train UNITE. Notably, to reduce the gap between pre-training and fine-tuning, we use data cropping and a ranking-based score normalization strategy. During the fine-tuning phase, we use both Direct Assessment (DA) and Multidimensional Quality Metrics (MQM) data from past years' WMT competitions. Specially, we collect the results from models with different pre-trained language model backbones, and use different ensembling strategies for involved translation directions.
How reliably can we trust the scores obtained from social bias benchmarks as faithful indicators of problematic social biases in a given language model? In this work, we study this question by contrasting social biases with non-social biases stemming from choices made during dataset construction that might not even be discernible to the human eye. To do so, we empirically simulate various alternative constructions for a given benchmark based on innocuous modifications (such as paraphrasing or random-sampling) that maintain the essence of their social bias. On two well-known social bias benchmarks (Winogender and BiasNLI) we observe that these shallow modifications have a surprising effect on the resulting degree of bias across various models. We hope these troubling observations motivate more robust measures of social biases.
Pretraining a language model (LM) on text has been shown to help various downstream NLP tasks. Recent works show that a knowledge graph (KG) can complement text data, offering structured background knowledge that provides a useful scaffold for reasoning. However, these works are not pretrained to learn a deep fusion of the two modalities at scale, limiting the potential to acquire fully joint representations of text and KG. Here we propose DRAGON (Deep Bidirectional Language-Knowledge Graph Pretraining), a self-supervised approach to pretraining a deeply joint language-knowledge foundation model from text and KG at scale. Specifically, our model takes pairs of text segments and relevant KG subgraphs as input and bidirectionally fuses information from both modalities. We pretrain this model by unifying two self-supervised reasoning tasks, masked language modeling and KG link prediction. DRAGON outperforms existing LM and LM+KG models on diverse downstream tasks including question answering across general and biomedical domains, with +5% absolute gain on average. In particular, DRAGON achieves notable performance on complex reasoning about language and knowledge (+10% on questions involving long contexts or multi-step reasoning) and low-resource QA (+8% on OBQA and RiddleSense), and new state-of-the-art results on various BioNLP tasks. Our code and trained models are available at https://github.com/michiyasunaga/dragon.
Knowledge-enhanced Pre-trained Language Model (PLM) has recently received significant attention, which aims to incorporate factual knowledge into PLMs. However, most existing methods modify the internal structures of fixed types of PLMs by stacking complicated modules, and introduce redundant and irrelevant factual knowledge from knowledge bases (KBs). In this paper, to address these problems, we introduce a seminal knowledge prompting paradigm and further propose a knowledge-prompting-based PLM framework KP-PLM. This framework can be flexibly combined with existing mainstream PLMs. Specifically, we first construct a knowledge sub-graph from KBs for each context. Then we design multiple continuous prompts rules and transform the knowledge sub-graph into natural language prompts. To further leverage the factual knowledge from these prompts, we propose two novel knowledge-aware self-supervised tasks including prompt relevance inspection and masked prompt modeling. Extensive experiments on multiple natural language understanding (NLU) tasks show the superiority of KP-PLM over other state-of-the-art methods in both full-resource and low-resource settings.
Speakers repeat constructions frequently in dialogue. Due to their peculiar information-theoretic properties, repetitions can be thought of as a strategy for cost-effective communication. In this study, we focus on the repetition of lexicalised constructions -- i.e., recurring multi-word units -- in English open-domain spoken dialogues. We hypothesise that speakers use construction repetition to mitigate information rate, leading to an overall decrease in utterance information content over the course of a dialogue. We conduct a quantitative analysis, measuring the information content of constructions and that of their containing utterances, estimating information content with an adaptive neural language model. We observe that construction usage lowers the information content of utterances. This facilitating effect (i) increases throughout dialogues, (ii) is boosted by repetition, (iii) grows as a function of repetition frequency and density, and (iv) is stronger for repetitions of referential constructions.
Language models (LMs) now excel at many tasks such as few-shot learning, question answering, reasoning, and dialog. However, they sometimes generate unsupported or misleading content. A user cannot easily determine whether their outputs are trustworthy or not, because most LMs do not have any built-in mechanism for attribution to external evidence. To enable attribution while still preserving all the powerful advantages of recent generation models, we propose RARR (Retrofit Attribution using Research and Revision), a system that 1) automatically finds attribution for the output of any text generation model and 2) post-edits the output to fix unsupported content while preserving the original output as much as possible. When applied to the output of several state-of-the-art LMs on a diverse set of generation tasks, we find that RARR significantly improves attribution while otherwise preserving the original input to a much greater degree than previously explored edit models. Furthermore, the implementation of RARR requires only a handful of training examples, a large language model, and standard web search.
Norm discovery is important for understanding and reasoning about the acceptable behaviors and potential violations in human communication and interactions. We introduce NormSage, a framework for addressing the novel task of conversation-grounded multi-lingual, multi-cultural norm discovery, based on language model prompting and self-verification. NormSAGE leverages the expressiveness and implicit knowledge of the pretrained GPT-3 language model backbone, to elicit knowledge about norms through directed questions representing the norm discovery task and conversation context. It further addresses the risk of language model hallucination with a self-verification mechanism ensuring that the norms discovered are correct and are substantially grounded to their source conversations. Evaluation results show that our approach discovers significantly more relevant and insightful norms for conversations on-the-fly compared to baselines (>10+% in Likert scale rating). The norms discovered from Chinese conversation are also comparable to the norms discovered from English conversation in terms of insightfulness and correctness (<3% difference). In addition, the culture-specific norms are promising quality, allowing for 80% accuracy in culture pair human identification. Finally, our grounding process in norm discovery self-verification can be extended for instantiating the adherence and violation of any norm for a given conversation on-the-fly, with explainability and transparency. NormSAGE achieves an AUC of 95.4% in grounding, with natural language explanation matching human-written quality.
We present M2D2, a fine-grained, massively multi-domain corpus for studying domain adaptation in language models (LMs). M2D2 consists of 8.5B tokens and spans 145 domains extracted from Wikipedia and Semantic Scholar. Using ontologies derived from Wikipedia and ArXiv categories, we organize the domains in each data source into 22 groups. This two-level hierarchy enables the study of relationships between domains and their effects on in- and out-of-domain performance after adaptation. We also present a number of insights into the nature of effective domain adaptation in LMs, as examples of the new types of studies M2D2 enables. To improve in-domain performance, we show the benefits of adapting the LM along a domain hierarchy; adapting to smaller amounts of fine-grained domain-specific data can lead to larger in-domain performance gains than larger amounts of weakly relevant data. We further demonstrate a trade-off between in-domain specialization and out-of-domain generalization within and across ontologies, as well as a strong correlation between out-of-domain performance and lexical overlap between domains.
Demographic factors (e.g., gender or age) shape our language. Previous work showed that incorporating demographic factors can consistently improve performance for various NLP tasks with traditional NLP models. In this work, we investigate whether these previous findings still hold with state-of-the-art pretrained Transformer-based language models (PLMs). We use three common specialization methods proven effective for incorporating external knowledge into pretrained Transformers (e.g., domain-specific or geographic knowledge). We adapt the language representations for the demographic dimensions of gender and age, using continuous language modeling and dynamic multi-task learning for adaptation, where we couple language modeling objectives with the prediction of demographic classes. Our results, when employing a multilingual PLM, show substantial gains in task performance across four languages (English, German, French, and Danish), which is consistent with the results of previous work. However, controlling for confounding factors - primarily domain and language proficiency of Transformer-based PLMs - shows that downstream performance gains from our demographic adaptation do not actually stem from demographic knowledge. Our results indicate that demographic specialization of PLMs, while holding promise for positive societal impact, still represents an unsolved problem for (modern) NLP.
A common limitation of diagnostic tests for detecting social biases in NLP models is that they may only detect stereotypic associations that are pre-specified by the designer of the test. Since enumerating all possible problematic associations is infeasible, it is likely these tests fail to detect biases that are present in a model but not pre-specified by the designer. To address this limitation, we propose SODAPOP (SOcial bias Discovery from Answers about PeOPle) in social commonsense question-answering. Our pipeline generates modified instances from the Social IQa dataset (Sap et al., 2019) by (1) substituting names associated with different demographic groups, and (2) generating many distractor answers from a masked language model. By using a social commonsense model to score the generated distractors, we are able to uncover the model's stereotypic associations between demographic groups and an open set of words. We also test SODAPOP on debiased models and show the limitations of multiple state-of-the-art debiasing algorithms.
Transformer has achieved remarkable success in language, image, and speech processing. Recently, various efficient attention architectures have been proposed to improve transformer's efficiency while largely preserving its efficacy, especially in modeling long sequences. A widely-used benchmark to test these efficient methods' capability on long-range modeling is Long Range Arena (LRA). However, LRA only focuses on the standard bidirectional (or noncausal) self attention, and completely ignores cross attentions and unidirectional (or causal) attentions, which are equally important to downstream applications. In this paper, we propose Comprehensive Attention Benchmark (CAB) under a fine-grained attention taxonomy with four distinguishable attention patterns, namely, noncausal self, causal self, noncausal cross, and causal cross attentions. CAB collects seven real-world tasks from different research areas to evaluate efficient attentions under the four attention patterns. Among these tasks, CAB validates efficient attentions in eight backbone networks to show their generalization across neural architectures. We conduct exhaustive experiments to benchmark the performances of nine widely-used efficient attention architectures designed with different philosophies on CAB. Extensive experimental results also shed light on the fundamental problems of efficient attentions, such as efficiency length against vanilla attention, performance consistency across attention patterns, the benefit of attention mechanisms, and interpolation/extrapolation on long-context language modeling.
Structured knowledge is important for many AI applications. Commonsense knowledge, which is crucial for robust human-centric AI, is covered by a small number of structured knowledge projects. However, they lack knowledge about human traits and behaviors conditioned on socio-cultural contexts, which is crucial for situative AI. This paper presents CANDLE, an end-to-end methodology for extracting high-quality cultural commonsense knowledge (CCSK) at scale. CANDLE extracts CCSK assertions from a huge web corpus and organizes them into coherent clusters, for 3 domains of subjects (geography, religion, occupation) and several cultural facets (food, drinks, clothing, traditions, rituals, behaviors). CANDLE includes judicious techniques for classification-based filtering and scoring of interestingness. Experimental evaluations show the superiority of the CANDLE CCSK collection over prior works, and an extrinsic use case demonstrates the benefits of CCSK for the GPT-3 language model. Code and data can be accessed at https://candle.mpi-inf.mpg.de/.
Automatic evaluation metrics are crucial to the development of generative systems. In recent years, pre-trained language model (PLM) based metrics, such as BERTScore, have been commonly adopted in various generation tasks. However, it has been demonstrated that PLMs encode a range of stereotypical societal biases, leading to a concern on the fairness of PLMs as metrics. To that end, this work presents the first systematic study on the social bias in PLM-based metrics. We demonstrate that popular PLM-based metrics exhibit significantly higher social bias than traditional metrics on 6 sensitive attributes, namely race, gender, religion, physical appearance, age, and socioeconomic status. In-depth analysis suggests that choosing paradigms (matching, regression, or generation) of the metric has a greater impact on fairness than choosing PLMs. In addition, we develop debiasing adapters that are injected into PLM layers, mitigating bias in PLM-based metrics while retaining high performance for evaluating text generation.