Dialog response generation in open domain is an important research topic where the main challenge is to generate relevant and diverse responses. In this paper, we propose a new dialog pre-training framework called DialogVED, which introduces continuous latent variables into the enhanced encoder-decoder pre-training framework to increase the relevance and diversity of responses. With the help of a large dialog corpus (Reddit), we pre-train the model using the following 4 tasks adopted in language models (LMs) and variational autoencoders (VAEs): 1) masked language model; 2) response generation; 3) bag-of-words prediction; and 4) KL divergence reduction. We also add additional parameters to model the turn structure in dialogs to improve the performance of the pre-trained model. We conduct experiments on PersonaChat, DailyDialog, and DSTC7-AVSD benchmarks for response generation. Experimental results show that our model achieves the new state-of-the-art results on all these datasets.
Code summarization with deep learning has been widely studied in recent years. Current deep learning models for code summarization generally follow the principle in neural machine translation and adopt the encoder-decoder framework, where the encoder learns the semantic representations from source code and the decoder transforms the learnt representations into human-readable text that describes the functionality of code snippets. Despite they achieve the new state-of-the-art performance, we notice that current models often either generate less fluent summaries, or fail to capture the core functionality, since they usually focus on a single type of code representations. As such we propose GypSum, a new deep learning model that learns hybrid representations using graph attention neural networks and a pre-trained programming and natural language model. We introduce particular edges related to the control flow of a code snippet into the abstract syntax tree for graph construction, and design two encoders to learn from the graph and the token sequence of source code, respectively. We modify the encoder-decoder sublayer in the Transformer's decoder to fuse the representations and propose a dual-copy mechanism to facilitate summary generation. Experimental results demonstrate the superior performance of GypSum over existing code summarization models.
Being a popular mode of text-based communication in multilingual communities, code-mixing in online social media has became an important subject to study. Learning the semantics and morphology of code-mixed language remains a key challenge, due to scarcity of data and unavailability of robust and language-invariant representation learning technique. Any morphologically-rich language can benefit from character, subword, and word-level embeddings, aiding in learning meaningful correlations. In this paper, we explore a hierarchical transformer-based architecture (HIT) to learn the semantics of code-mixed languages. HIT consists of multi-headed self-attention and outer product attention components to simultaneously comprehend the semantic and syntactic structures of code-mixed texts. We evaluate the proposed method across 6 Indian languages (Bengali, Gujarati, Hindi, Tamil, Telugu and Malayalam) and Spanish for 9 NLP tasks on 17 datasets. The HIT model outperforms state-of-the-art code-mixed representation learning and multilingual language models in all tasks. We further demonstrate the generalizability of the HIT architecture using masked language modeling-based pre-training, zero-shot learning, and transfer learning approaches. Our empirical results show that the pre-training objectives significantly improve the performance on downstream tasks.
Machine translation models struggle when translating out-of-domain text, which makes domain adaptation a topic of critical importance. However, most domain adaptation methods focus on fine-tuning or training the entire or part of the model on every new domain, which can be costly. On the other hand, semi-parametric models have been shown to successfully perform domain adaptation by retrieving examples from an in-domain datastore (Khandelwal et al., 2021). A drawback of these retrieval-augmented models, however, is that they tend to be substantially slower. In this paper, we explore several approaches to speed up nearest neighbor machine translation. We adapt the methods recently proposed by He et al. (2021) for language modeling, and introduce a simple but effective caching strategy that avoids performing retrieval when similar contexts have been seen before. Translation quality and runtimes for several domains show the effectiveness of the proposed solutions.
Simile interpretation (SI) and simile generation (SG) are challenging tasks for NLP because models require adequate world knowledge to produce predictions. Previous works have employed many hand-crafted resources to bring knowledge-related into models, which is time-consuming and labor-intensive. In recent years, pre-trained language models (PLMs) based approaches have become the de-facto standard in NLP since they learn generic knowledge from a large corpus. The knowledge embedded in PLMs may be useful for SI and SG tasks. Nevertheless, there are few works to explore it. In this paper, we probe simile knowledge from PLMs to solve the SI and SG tasks in the unified framework of simile triple completion for the first time. The backbone of our framework is to construct masked sentences with manual patterns and then predict the candidate words in the masked position. In this framework, we adopt a secondary training process (Adjective-Noun mask Training) with the masked language model (MLM) loss to enhance the prediction diversity of candidate words in the masked position. Moreover, pattern ensemble (PE) and pattern search (PS) are applied to improve the quality of predicted words. Finally, automatic and human evaluations demonstrate the effectiveness of our framework in both SI and SG tasks.
The UMLS Metathesaurus integrates more than 200 biomedical source vocabularies. During the Metathesaurus construction process, synonymous terms are clustered into concepts by human editors, assisted by lexical similarity algorithms. This process is error-prone and time-consuming. Recently, a deep learning model (LexLM) has been developed for the UMLS Vocabulary Alignment (UVA) task. This work introduces UBERT, a BERT-based language model, pretrained on UMLS terms via a supervised Synonymy Prediction (SP) task replacing the original Next Sentence Prediction (NSP) task. The effectiveness of UBERT for UMLS Metathesaurus construction process is evaluated using the UMLS Vocabulary Alignment (UVA) task. We show that UBERT outperforms the LexLM, as well as biomedical BERT-based models. Key to the performance of UBERT are the synonymy prediction task specifically developed for UBERT, the tight alignment of training data to the UVA task, and the similarity of the models used for pretrained UBERT.
Modern day conversational agents are trained to emulate the manner in which humans communicate. To emotionally bond with the user, these virtual agents need to be aware of the affective state of the user. Transformers are the recent state of the art in sequence-to-sequence learning that involves training an encoder-decoder model with word embeddings from utterance-response pairs. We propose an emotion-aware transformer encoder for capturing the emotional quotient in the user utterance in order to generate human-like empathetic responses. The contributions of our paper are as follows: 1) An emotion detector module trained on the input utterances determines the affective state of the user in the initial phase 2) A novel transformer encoder is proposed that adds and normalizes the word embedding with emotion embedding thereby integrating the semantic and affective aspects of the input utterance 3) The encoder and decoder stacks belong to the Transformer-XL architecture which is the recent state of the art in language modeling. Experimentation on the benchmark Facebook AI empathetic dialogue dataset confirms the efficacy of our model from the higher BLEU-4 scores achieved for the generated responses as compared to existing methods. Emotionally intelligent virtual agents are now a reality and inclusion of affect as a modality in all human-machine interfaces is foreseen in the immediate future.
Despite the importance of understanding causality, corpora addressing causal relations are limited. There is a discrepancy between existing annotation guidelines of event causality and conventional causality corpora that focus more on linguistics. Many guidelines restrict themselves to include only explicit relations or clause-based arguments. Therefore, we propose an annotation schema for event causality that addresses these concerns. We annotated 3,559 event sentences from protest event news with labels on whether it contains causal relations or not. Our corpus is known as the Causal News Corpus (CNC). A neural network built upon a state-of-the-art pre-trained language model performed well with 81.20% F1 score on test set, and 83.46% in 5-folds cross-validation. CNC is transferable across two external corpora: CausalTimeBank (CTB) and Penn Discourse Treebank (PDTB). Leveraging each of these external datasets for training, we achieved up to approximately 64% F1 on the CNC test set without additional fine-tuning. CNC also served as an effective training and pre-training dataset for the two external corpora. Lastly, we demonstrate the difficulty of our task to the layman in a crowd-sourced annotation exercise. Our annotated corpus is publicly available, providing a valuable resource for causal text mining researchers.
Language models generate texts by successively predicting probability distributions for next tokens given past ones. A growing field of interest tries to leverage external information in the decoding process so that the generated texts have desired properties, such as being more natural, non toxic, faithful, or having a specific writing style. A solution is to use a classifier at each generation step, resulting in a cooperative environment where the classifier guides the decoding of the language model distribution towards relevant texts for the task at hand. In this paper, we examine three families of (transformer-based) discriminators for this specific task of cooperative decoding: bidirectional, left-to-right and generative ones. We evaluate the pros and cons of these different types of discriminators for cooperative generation, exploring respective accuracy on classification tasks along with their impact on the resulting sample quality and computational performances. We also provide the code of a batched implementation of the powerful cooperative decoding strategy used for our experiments, the Monte Carlo Tree Search, working with each discriminator for Natural Language Generation.
Pre-trained language models (PLMs) have achieved remarkable success on various natural language understanding tasks. Simple fine-tuning of PLMs, on the other hand, might be suboptimal for domain-specific tasks because they cannot possibly cover knowledge from all domains. While adaptive pre-training of PLMs can help them obtain domain-specific knowledge, it requires a large training cost. Moreover, adaptive pre-training can harm the PLM's performance on the downstream task by causing catastrophic forgetting of its general knowledge. To overcome such limitations of adaptive pre-training for PLM adaption, we propose a novel domain adaption framework for PLMs coined as Knowledge-Augmented Language model Adaptation (KALA), which modulates the intermediate hidden representations of PLMs with domain knowledge, consisting of entities and their relational facts. We validate the performance of our KALA on question answering and named entity recognition tasks on multiple datasets across various domains. The results show that, despite being computationally efficient, our KALA largely outperforms adaptive pre-training. Code is available at: https://github.com/Nardien/KALA/.
Ranking responses for a given dialogue context is a popular benchmark in which the setup is to re-rank the ground-truth response over a limited set of $n$ responses, where $n$ is typically 10. The predominance of this setup in conversation response ranking has lead to a great deal of attention to building neural re-rankers, while the first-stage retrieval step has been overlooked. Since the correct answer is always available in the candidate list of $n$ responses, this artificial evaluation setup assumes that there is a first-stage retrieval step which is always able to rank the correct response in its top-$n$ list. In this paper we focus on the more realistic task of full-rank retrieval of responses, where $n$ can be up to millions of responses. We investigate both dialogue context and response expansion techniques for sparse retrieval, as well as zero-shot and fine-tuned dense retrieval approaches. Our findings based on three different information-seeking dialogue datasets reveal that a learned response expansion technique is a solid baseline for sparse retrieval. We find the best performing method overall to be dense retrieval with intermediate training, i.e. a step after the language model pre-training where sentence representations are learned, followed by fine-tuning on the target conversational data. We also investigate the intriguing phenomena that harder negatives sampling techniques lead to worse results for the fine-tuned dense retrieval models. The code and datasets are available at https://github.com/Guzpenha/transformer_rankers/tree/full_rank_retrieval_dialogues.
We propose DiffCSE, an unsupervised contrastive learning framework for learning sentence embeddings. DiffCSE learns sentence embeddings that are sensitive to the difference between the original sentence and an edited sentence, where the edited sentence is obtained by stochastically masking out the original sentence and then sampling from a masked language model. We show that DiffSCE is an instance of equivariant contrastive learning (Dangovski et al., 2021), which generalizes contrastive learning and learns representations that are insensitive to certain types of augmentations and sensitive to other "harmful" types of augmentations. Our experiments show that DiffCSE achieves state-of-the-art results among unsupervised sentence representation learning methods, outperforming unsupervised SimCSE by 2.3 absolute points on semantic textual similarity tasks.
Multi-modal data abounds in biomedicine, such as radiology images and reports. Interpreting this data at scale is essential for improving clinical care and accelerating clinical research. Biomedical text with its complex semantics poses additional challenges in vision--language modelling compared to the general domain, and previous work has used insufficiently adapted models that lack domain-specific language understanding. In this paper, we show that principled textual semantic modelling can substantially improve contrastive learning in self-supervised vision--language processing. We release a language model that achieves state-of-the-art results in radiology natural language inference through its improved vocabulary and novel language pretraining objective leveraging semantics and discourse characteristics in radiology reports. Further, we propose a self-supervised joint vision--language approach with a focus on better text modelling. It establishes new state of the art results on a wide range of publicly available benchmarks, in part by leveraging our new domain-specific language model. We release a new dataset with locally-aligned phrase grounding annotations by radiologists to facilitate the study of complex semantic modelling in biomedical vision--language processing. A broad evaluation, including on this new dataset, shows that our contrastive learning approach, aided by textual-semantic modelling, outperforms prior methods in segmentation tasks, despite only using a global-alignment objective.
Recent causal probing literature reveals when language models and syntactic probes use similar representations. Such techniques may yield "false negative" causality results: models may use representations of syntax, but probes may have learned to use redundant encodings of the same syntactic information. We demonstrate that models do encode syntactic information redundantly and introduce a new probe design that guides probes to consider all syntactic information present in embeddings. Using these probes, we find evidence for the use of syntax in models where prior methods did not, allowing us to boost model performance by injecting syntactic information into representations.
Sparse mixture of experts provides larger model capacity while requiring a constant computational overhead. It employs the routing mechanism to distribute input tokens to the best-matched experts according to their hidden representations. However, learning such a routing mechanism encourages token clustering around expert centroids, implying a trend toward representation collapse. In this work, we propose to estimate the routing scores between tokens and experts on a low-dimensional hypersphere. We conduct extensive experiments on cross-lingual language model pre-training and fine-tuning on downstream tasks. Experimental results across seven multilingual benchmarks show that our method achieves consistent gains. We also present a comprehensive analysis on the representation and routing behaviors of our models. Our method alleviates the representation collapse issue and achieves more consistent routing than the baseline mixture-of-experts methods.
Task-oriented dialogue generation is challenging since the underlying knowledge is often dynamic and effectively incorporating knowledge into the learning process is hard. It is particularly challenging to generate both human-like and informative responses in this setting. Recent research primarily focused on various knowledge distillation methods where the underlying relationship between the facts in a knowledge base is not effectively captured. In this paper, we go one step further and demonstrate how the structural information of a knowledge graph can improve the system's inference capabilities. Specifically, we propose DialoKG, a novel task-oriented dialogue system that effectively incorporates knowledge into a language model. Our proposed system views relational knowledge as a knowledge graph and introduces (1) a structure-aware knowledge embedding technique, and (2) a knowledge graph-weighted attention masking strategy to facilitate the system selecting relevant information during the dialogue generation. An empirical evaluation demonstrates the effectiveness of DialoKG over state-of-the-art methods on several standard benchmark datasets.
Knowledge distillation is an approach to transfer information on representations from a teacher to a student by reducing their difference. A challenge of this approach is to reduce the flexibility of the student's representations inducing inaccurate learning of the teacher's knowledge. To resolve it in transferring, we investigate distillation of structures of representations specified to three types: intra-feature, local inter-feature, global inter-feature structures. To transfer them, we introduce feature structure distillation methods based on the Centered Kernel Alignment, which assigns a consistent value to similar features structures and reveals more informative relations. In particular, a memory-augmented transfer method with clustering is implemented for the global structures. The methods are empirically analyzed on the nine tasks for language understanding of the GLUE dataset with Bidirectional Encoder Representations from Transformers (BERT), which is a representative neural language model. In the results, the proposed methods effectively transfer the three types of structures and improve performance compared to state-of-the-art distillation methods. Indeed, the code for the methods is available in https://github.com/maroo-sky/FSD.
Aspect-based sentiment analysis (ABSA) is a fine-grained sentiment analysis task, which focuses on detecting the sentiment polarity towards the aspect in a sentence. However, it is always sensitive to the multi-aspect challenge, where features of multiple aspects in a sentence will affect each other. To mitigate this issue, we design a novel training framework, called Contrastive Cross-Channel Data Augmentation (C3 DA), which leverages an in-domain generator to construct more multi-aspect samples and then boosts the robustness of ABSA models via contrastive learning on these generated data. In practice, given a generative pretrained language model and some limited ABSA labeled data, we first employ some parameter-efficient approaches to perform the in-domain fine-tuning. Then, the obtained in-domain generator is used to generate the synthetic sentences from two channels, i.e., Aspect Augmentation Channel and Polarity Augmentation Channel, which generate the sentence condition on a given aspect and polarity respectively. Specifically, our C3 DA performs the sentence generation in a cross-channel manner to obtain more sentences, and proposes an Entropy-Minimization Filter to filter low-quality generated samples. Extensive experiments show that our C3 DA can outperform those baselines without any augmentations by about 1% on accuracy and Macro- F1. Code and data are released in https://github.com/wangbing1416/C3DA.
Entity Set Expansion (ESE) is a promising task which aims to expand entities of the target semantic class described by a small seed entity set. Various NLP and IR applications will benefit from ESE due to its ability to discover knowledge. Although previous ESE methods have achieved great progress, most of them still lack the ability to handle hard negative entities (i.e., entities that are difficult to distinguish from the target entities), since two entities may or may not belong to the same semantic class based on different granularity levels we analyze on. To address this challenge, we devise an entity-level masked language model with contrastive learning to refine the representation of entities. In addition, we propose the ProbExpan, a novel probabilistic ESE framework utilizing the entity representation obtained by the aforementioned language model to expand entities. Extensive experiments and detailed analyses on three datasets show that our method outperforms previous state-of-the-art methods.
This paper describes the BLCU-ICALL system used in the SemEval-2022 Task 1 Comparing Dictionaries and Word Embeddings, the Definition Modeling subtrack, achieving 1st on Italian, 2nd on Spanish and Russian, and 3rd on English and French. We propose a transformer-based multitasking framework to explore the task. The framework integrates multiple embedding architectures through the cross-attention mechanism, and captures the structure of glosses through a masking language model objective. Additionally, we also investigate a simple but effective model ensembling strategy to further improve the robustness. The evaluation results show the effectiveness of our solution. We release our code at: https://github.com/blcuicall/SemEval2022-Task1-DM.