Articles

Knowledge selection is an important and chal- lenging task which could provide the appropri- ate knowledge for informative dialogue gen- eration. However, the needed gold knowl- edge label is difficult to collect in reality. In this paper, we study knowledge selection for dialogue generation in the unsupervised sce- nario and …...

Time-to-event prediction tasks are common in conversation modelling, for applications such as predicting the length of a conversation or when a user will stop contributing to a plat- form. Despite the fact that it is natural to frame such predictions as regression tasks, re- cent work has modelled them as …...

In multi-label text classification (MLTC), each given document is associated with a set of correlated labels. To capture label correla- tions, previous classifier-chain and sequence- to-sequence models transform MLTC to a se- quence prediction task. However, they tend to suffer from label order dependency, la- bel combination over-fitting and error …...

Recent years pretrained language models (PLMs) hit a success on several downstream tasks, showing their power on modeling lan- guage. To better understand and leverage what PLMs have learned, several techniques have emerged to explore syntactic structures entailed by PLMs. However, few efforts have been made to explore grounding capabilities …...

Transferring knowledge from a label-rich do- main (source domain) to a label-scarce domain (target domain) for pervasive cross-domain Text Classification (TC) is a non-trivial task. To overcome this issue, we propose EADA, a novel unsupervised energy-based adversarial domain adaptation framework. First, a deep pre-trained language model (e.g. RoBERTa) is leveraged …...

Commonsense question answering (QA) re- quires a model to grasp commonsense and factual knowledge to answer questions about world events. Many prior methods couple lan- guage modeling with knowledge graphs (KG). However, although a KG contains rich struc- tural information, it lacks the context to pro- vide a more precise …...

This work studies the long-standing prob- lems of model capacity and negative interfer- ence in multilingual neural machine transla- tion (MNMT). We use network pruning tech- niques and observe that pruning 50-70% of the parameters from a trained MNMT model results only in a 0.29-1.98 drop in the BLEU score. …...

Do state-of-the-art natural language under- standing models care about word order? Not always! We found 75% to 90% of the correct predictions of BERT-based classifiers, trained on many GLUE tasks, remain constant after in- put words are randomly shuffled. Although BERT embeddings are famously contextual, the contribution of each individual …...

Neural sequence models exhibit limited com- positional generalization ability in semantic parsing tasks. Compositional generalization requires algebraic recombination, i.e., dynam- ically recombining structured expressions in a recursive manner. However, most previ- ous studies mainly concentrate on recombin- ing lexical units, which is an important but not sufficient part of algebraic …...

While research on explaining predictions of open-domain QA systems (ODQA) is gain- ing momentum, most works do not evaluate whether these explanations improve user trust. Furthermore, many users interact with ODQA using voice-assistants, yet prior works exclu- sively focus on visual displays, risking (as we also show) incorrectly extrapolating the …...

Semantic embedding has been widely investi- gated for aligning knowledge graph (KG) en- tities. Current methods have explored and uti- lized the graph structure, the entity names, and attributes, but ignore the ontology (or ontologi- cal schema) which contains critical meta infor- mation such as classes and their membership relationships …...

In reinforcement learning (RL) based task- oriented dialogue systems, users act as the en- vironment and the agent learns the policy by interacting with users. However, due to the subjectivity of different users, the complexity of user-generated training conversations varies greatly, which leads to different difficulties for the agent to …...

We study the possibilities of building a non-autoregressive speech-to-text translation model using connectionist temporal classifi- cation (CTC), and use CTC-based automatic speech recognition as an auxiliary task to im- prove the performance. CTC’s success on translation is counter-intuitive due to its mono- tonicity assumption, so we analyze its reorder- ing …...

Code summarization (CS) is becoming a promising area in recent language understand- ing, which aims to generate sensible human language automatically for programming lan- guage in the format of source code, serving in the most convenience of programmer de- veloping. It is well known that program- ming languages are highly …...

A good open-domain chatbot should avoid pre- senting contradictory responses about facts or opinions in a conversational session, known as its consistency capacity. However, evalu- ating the consistency capacity of a chatbot is still challenging. Employing human judges to interact with chatbots on purpose to check their capacities is costly …...

Knowledge retrieval and reasoning are two key stages in multi-hop question answering (QA) at web scale. Existing approaches suf- fer from low confidence when retrieving evi- dence facts to fill the knowledge gap and lack transparent reasoning process. In this paper, we propose a new framework to exploit more valid …...

Medical code assignment from clinical text is a fundamental task in clinical information system management. As medical notes are typically lengthy and the medical coding sys- tem’s code space is large, this task is a long- standing challenge. Recent work applies deep neural network models to encode the medical notes …...

We propose a simple and effective few-shot model for slot tagging. Recent work shows that it is promising to extend standard few- shot classification methods to sequence label- ing with CRF-specific augmentations. Such methods show strengths in encoding slot name semantics and slot dependencies. However, we find these strengths can …...

Sensor names as alphanumeric strings typi- cally encode their key contextual information such as their function or physical location. We focus here on sensors used in smart build- ing applications. In these applications, sen- sor names are curated in a building vendor- specific manner using different structures and esoteric vocabularies. …...

While word segmentation is a solved prob- lem in many languages, it is still a chal- lenge in continuous-script or low-resource lan- guages. Like other NLP tasks, word segmen- tation is domain-dependent, which can be a challenge in low-resource languages like Thai and Urdu since there can be domains with …...