Non-autoregressive (NAR) models have been mainly developed to improve decoding efficiency. Lately, they have also shown great potential in controlled text generation tasks. In this work, we investigate the decoding space of NAR models through lexically constrained machine translation tasks, and develop a search-based decoding algorithm named LexMAP, which is comparable to the autoregressive Grid Beam Search (GBS) method. Our analysis reveals several interesting properties of NAR decoding: 1) the NAR-based method does not suffer from the MAP degradation issue as the autoregressive method does; 2) AR beam search exhibits strong positional bias, in which the candidates only diverge at the end of the sequence; 3) NAR search explores a larger portion of the probability space, suggesting that the search algorithm better exploits the model’s potential.

Knowledge graph-based dialogue generation (KG-DG) is a challenging task requiring models to effectively incorporate external knowledge into conversational responses. While large language models (LLMs) have achieved impressive results across various NLP tasks, their ability to utilize external knowledge in KG-DG remains under-explored. We observe that LLMs often rely on internal knowledge, leading to detachment from provided knowledge graphs, even when they are given a flawlessly retrieved knowledge graph. First, we introduce LLM-KAT, an evaluation procedure for measuring knowledge attachment in generated responses. Second, we propose a simple yet effective entity anonymization technique to encourage LLMs to better leverage external knowledge. Experiments on the OpenDialKG dataset demonstrate that our approach improves LLMs' attachment on external knowledge.

The ability of zero-shot translation emerges when we train a multilingual model with certain translation directions; the model can then directly translate in unseen directions. Alternatively, zero-shot translation can be accomplished by pivoting through a third language (e.g., English). In our work, we observe that both direct and pivot translations are noisy and achieve less satisfactory performance. We propose EBBS, an ensemble method with a novel bi-level beam search algorithm, where each ensemble component explores its own prediction step by step at the lower level but all components are synchronized by a "soft voting" mechanism at the upper level. Results on two popular multilingual translation datasets show that EBBS consistently outperforms direct and pivot translations, as well as existing ensemble techniques. Further, we can distill the ensemble's knowledge back to the multilingual model to improve inference efficiency; profoundly, our EBBS-distilled model can even outperform EBBS as it learns from the ensemble knowledge.

Length-control summarization aims to condense long texts into a short one within a certain length limit. Previous approaches often use autoregressive (AR) models and treat the length requirement as a soft constraint, which may not always be satisfied. In this study, we propose a novel length-control decoding algorithm based on the directed acyclic Transformer (DAT). Our approach allows for multiple plausible sequence fragments and predicts a path to connect them. In addition, we propose a Sequence Maximum a Posteriori (Seq-MAP) decoding algorithm that marginalizes different possible paths and finds the most probable summary satisfying the length budget. Our algorithm is based on beam search, which further facilitates a reranker for performance improvement. Experimental results on the Gigaword dataset demonstrate our state-of-the-art performance for length-control summarization.

The proliferation of social media platforms has given rise to the amount of online debates and arguments. Consequently, the need for automatic summarization methods for such debates is imperative, however this area of summarization is rather understudied. The Key Point Analysis (KPA) task formulates argument summarization as representing the summary of a large collection of arguments in the form of concise sentences in bullet-style format, called key points. A sub-task of KPA, called Key Point Generation (KPG), focuses on generating these key points given the arguments. This paper introduces a novel extractive approach for key point generation, that outperforms previous state-of-the-art methods for the task. Our method utilizes an extractive clustering based approach that offers concise, high quality generated key points with higher coverage of reference summaries, and less redundant outputs. In addition, we show that the existing evaluation metrics for summarization such as ROUGE are incapable of differentiating between generated key points of different qualities. To this end, we propose a new evaluation metric for assessing the generated key points by their coverage. Our code can be accessed online.

Recently, there has been considerable attention on detecting hallucinations and omissions in Machine Translation (MT) systems. The two dominant approaches to tackle this task involve analyzing the MT system's internal states or relying on the output of external tools, such as sentence similarity or MT quality estimators. In this work, we introduce OTTAWA, a novel Optimal Transport (OT)-based word aligner specifically designed to enhance the detection of hallucinations and omissions in MT systems. Our approach explicitly models the missing alignments by introducing a "null" vector, for which we propose a novel one-side constrained OT setting to allow an adaptive null alignment. Our approach yields competitive results compared to state-of-the-art methods across 18 language pairs on the HalOmi benchmark. In addition, it shows promising features, such as the ability to distinguish between both error types and perform word-level detection without accessing the MT system's internal states.

Overfitting, a common problem in Machine Learning, occurs when a predictive model learns the noise in the training data instead of the true underlying patterns and converges to perform very well with the training data but poorly with unseen data. Models that overfit cannot be deployed in practice. Regularization is a technique typically used to help a model better generalize. This is usually achieved by adding a penalty term in the loss function to discourage the model from fitting noise, making it more robust to noise in the data and, therefore more generalizable. One method of regularization is to take some of the concentration (called Smoothing Ratio (SR)) from the data sample’s ground truth label and distribute it uniformly among all the other labels during training. This method is called label smoothing and is a simple yet effective method to improve generalization. In this work, we explore what happens if we distribute the SR to the non-ground truth labels based on how closely they are related to the ground truth label, instead of uniformly. We call this approach of distributing the SR based on relation between labels as Preferential Label Smoothing (PLS). PLS represents a more unified approach of performing label smoothing. Ordinary uniform label smoothing becomes pointless as the number of labels becomes large since the SR proportion distributed per label becomes negligible. PLS is inconsequential in the case of binary classification, since there are only two labels. Therefore, we investigate the effects of PLS when the number of labels in the dataset is high. We also examine the effects of uniform and preferential label smoothing, as well as the absence of label smoothing, on the training dynamics. We demonstrate our study on image classification and text classification.

In this paper, we present empirical analysis on basic and depression specific multi-emotion mining in Tweets with the help of state of the art multi-label classifiers. We choose our basic emotions from a hybrid emotion model consisting of the common emotions from four highly regarded psychological models of emotions. Moreover, we augment that emotion model with new emotion categories because of their importance in the analysis of depression. Most of those additional emotions have not been used in previous emotion mining research. Our experimental analyses show that a cost sensitive RankSVM algorithm and a Deep Learning model are both robust, measured by both Macro F-measures and Micro F-measures. This suggests that these algorithms are superior in addressing the widely known data imbalance problem in multi-label learning. Moreover, our application of Deep Learning performs the best, giving it an edge in modeling deep semantic features of our extended emotional categories.

Multilingual neural machine translation (MNMT) aims at using one single model for multiple translation directions.Recent work applies non-autoregressive Transformers to improve the efficiency of MNMT, but requires expensive knowledge distillation (KD) processes.To this end, we propose an M-DAT approach to non-autoregressive multilingual machine translation.Our system leverages the recent advance of the directed acyclic Transformer (DAT), which does not require KD.We further propose a pivot back-translation (PivotBT) approach to improve the generalization to unseen translation directions.Experiments show that our M-DAT achieves state-of-the-art performance in non-autoregressive MNMT.

How do we perform efficient inference while retaining high translation quality? Existing neural machine translation models, such as Transformer, achieve high performance, but they decode words one by one, which is inefficient. Recent non-autoregressive translation models speed up the inference, but their quality is still inferior. In this work, we propose DSLP, a highly efficient and high-performance model for machine translation. The key insight is to train a non-autoregressive Transformer with Deep Supervision and feed additional Layer-wise Predictions. We conducted extensive experiments on four translation tasks (both directions of WMT'14 EN-DE and WMT'16 EN-RO). Results show that our approach consistently improves the BLEU scores compared with respective base models. Specifically, our best variant outperforms the autoregressive model on three translation tasks, while being 14.8 times more efficient in inference.

Text summarization aims to generate a short summary for an input text. In this work, we propose a Non-Autoregressive Unsupervised Summarization (NAUS) approach, which does not require parallel data for training. Our NAUS first performs edit-based search towards a heuristically defined score, and generates a summary as pseudo-groundtruth. Then, we train an encoder-only non-autoregressive Transformer based on the search result. We also propose a dynamic programming approach for length-control decoding, which is important for the summarization task. Experiments on two datasets show that NAUS achieves state-of-the-art performance for unsupervised summarization, yet largely improving inference efficiency. Further, our algorithm is able to perform explicit length-transfer summary generation.

Multi-label emotion classification is an important task in NLP and is essential to many applications. In this work, we propose a sequence-to-emotion (Seq2Emo) approach, which implicitly models emotion correlations in a bi-directional decoder. Experiments on SemEval’18 and GoEmotions datasets show that our approach outperforms state-of-the-art methods (without using external data). In particular, Seq2Emo outperforms the binary relevance (BR) and classifier chain (CC) approaches in a fair setting.

Although deep models have greatly improved the accuracy and robustness of image segmentation, obtaining segmentation results with highly accurate boundaries and fine structures is still a challenging problem. In this paper, we propose a simple yet powerful Boundary-Aware Segmentation Network (BASNet), which comprises a predict-refine architecture and a hybrid loss, for highly accurate image segmentation. The predict-refine architecture consists of a densely supervised encoder-decoder network and a residual refinement module, which are respectively used to predict and refine a segmentation probability map. The hybrid loss is a combination of the binary cross entropy, structural similarity and intersection-over-union losses, which guide the network to learn three-level (ie, pixel-, patch- and map- level) hierarchy representations. We evaluate our BASNet on two reverse tasks including salient object segmentation, camouflaged object segmentation, showing that it achieves very competitive performance with sharp segmentation boundaries. Importantly, BASNet runs at over 70 fps on a single GPU which benefits many potential real applications. Based on BASNet, we further developed two (close to) commercial applications: AR COPY & PASTE, in which BASNet is integrated with augmented reality for "COPYING" and "PASTING" real-world objects, and OBJECT CUT, which is a web-based tool for automatic object background removal. Both applications have already drawn huge amount of attention and have important real-world impacts. The code and two applications will be publicly available at: https://github.com/NathanUA/BASNet.

It is a common belief that training deep transformers from scratch requires large datasets. Consequently, for small datasets, people usually use shallow and simple additional layers on top of pre-trained models during fine-tuning. This work shows that this does not always need to be the case: with proper initialization and optimization, the benefits of very deep transformers can carry over to challenging tasks with small datasets, including Text-to-SQL semantic parsing and logical reading comprehension. In particular, we successfully train 48 layers of transformers, comprising 24 fine-tuned layers from pre-trained RoBERTa and 24 relation-aware layers trained from scratch. With fewer training steps and no task-specific pre-training, we obtain the state-of-the-art performance on the challenging cross-domain Text-to-SQL parsing benchmark Spider. We achieve this by deriving a novel Data-dependent Transformer Fixed-update initialization scheme (DT-Fixup), inspired by the prior T-Fixup work. Further error analysis shows that increasing depth can help improve generalization on small datasets for hard cases that require reasoning and structural understanding.

In this paper, we propose a globally normalized model for context-free grammar (CFG)-based semantic parsing. Instead of predicting a probability, our model predicts a real-valued score at each step and does not suffer from the label bias problem. Experiments show that our approach outperforms locally normalized models on small datasets, but it does not yield improvement on a large dataset.

Explicitly modeling emotions in dialogue generation has important applications, such as building empathetic personal companions. In this study, we consider the task of expressing a specific emotion for dialogue generation. Previous approaches take the emotion as a training signal, which may be ignored during inference. Here, we propose a search-based emotional dialogue system by simulated annealing (SA). Specifically, we first define a scoring function that combines contextual coherence and emotional correctness. Then, SA iteratively edits a general response, and search for a generation with a high score. In this way, we enforce the presence of the desired emotion. We evaluate our system on the NLPCC2017 dataset. The proposed method shows about 12% improvements in emotion accuracy compared with the previous state-of-the-art method, without hurting the generation quality (measured by BLEU).

In this paper, we describe our mUlti-task learNIng for cOmmonsense reasoNing (UNION) system submitted for Task C of the SemEval2020 Task 4, which is to generate a reason explaining why a given false statement is non-sensical. However, we found in the early experiments that simple adaptations such as fine-tuning GPT2 often yield dull and non-informative generations (e.g. simple negations). In order to generate more meaningful explanations, we propose UNION, a unified end-to-end framework, to utilize several existing commonsense datasets so that it allows a model to learn more dynamics under the scope of commonsense reasoning. In order to perform model selection efficiently, accurately and promptly, we also propose a couple of auxiliary automatic evaluation metrics so that we can extensively compare the models from different perspectives. Our submitted system not only results in a good performance in the proposed metrics but also outperforms its competitors with the highest achieved score of 2.10 for human evaluation while remaining a BLEU score of 15.7. Our code is made publicly available at GitHub.

•A novel ReSidual U-block (RSU) is designed to capture multi-scale deep features.•A nested U-structure, called U2-Net, that uses RSU is developed for salient object detection.•Both large (176.3 MB) and small (4.7 MB) instances of U2-Net get competitive results.

This paper describes the system submitted by ANA Team for the SemEval-2019 Task 3: EmoContext. We propose a novel Hierarchi- cal LSTMs for Contextual Emotion Detection (HRLCE) model. It classifies the emotion of an utterance given its conversational con- text. The results show that, in this task, our HRCLE outperforms the most recent state-of- the-art text classification framework: BERT. We combine the results generated by BERT and HRCLE to achieve an overall score of 0.7709 which ranked 5th on the final leader board of the competition among 165 Teams.