Grammatical Error Correction (GEC) is the task of correcting errorful
se...
Despite the great success of pre-trained language models, it is still a
...
Entailment Graphs (EGs) have been constructed based on extracted corpora...
We introduce MoviePuzzle, a novel challenge that targets visual narrativ...
The multi-answer phenomenon, where a question may have multiple answers
...
Causal reasoning, the ability to identify cause-and-effect relationship,...
We introduce CDBERT, a new learning paradigm that enhances the semantics...
Video-grounded dialogue understanding is a challenging problem that requ...
Event temporal relation extraction (ETRE) is usually formulated as a
mul...
Existing continual learning (CL) research regards catastrophic forgettin...
Existing research has shown that a multilingual pre-trained language mod...
Enhancing word usage is a desired feature for writing assistance. To fur...
Information retrieval (IR) plays a crucial role in locating relevant
res...
Position embeddings, encoding the positional relationships among tokens ...
Large Language Models (LLMs) have shown remarkable performance in variou...
Large pre-trained language models help to achieve state of the art on a
...
Although many large-scale knowledge bases (KBs) claim to contain multili...
To improve the performance of the dual-encoder retriever, one effective
...
Responding with multi-modal content has been recognized as an essential
...
The charge prediction task aims to predict the charge for a case given i...
We study video-grounded dialogue generation, where a response is generat...
People can acquire knowledge in an unsupervised manner by reading, and
c...
Multi-hop Knowledge Base Question Answering(KBQA) aims to find the answe...
DocRED is a widely used dataset for document-level relation extraction. ...
Typed entailment graphs try to learn the entailment relations between
pr...
Spatial commonsense, the knowledge about spatial position and relationsh...
In this paper, we present a new verification style reading comprehension...
Recent studies strive to incorporate various human rationales into neura...
Document-level Relation Extraction (RE) is a more challenging task than
...
Recent studies report that many machine reading comprehension (MRC) mode...
Causal inference is the process of capturing cause-effect relationship a...
Chinese pre-trained language models usually process text as a sequence o...
Recently, semantic parsing has attracted much attention in the community...
With the prosperous of cross-border e-commerce, there is an urgent deman...
Structural heterogeneity between knowledge graphs is an outstanding chal...
Text summarization is the research area aiming at creating a short and
c...
Stickers with vivid and engaging expressions are becoming increasingly
p...
Responding with knowledge has been recognized as an important capability...
Recent years have seen rapid progress in identifying predefined relation...
Sparsity is regarded as a desirable property of representations, especia...
Information-seeking conversation system aims at satisfying the informati...
Sponsored search optimizes revenue and relevance, which is estimated by
...
Existing dialog systems are all monolingual, where features shared among...
Text style transfer task requires the model to transfer a sentence of on...
Entity alignment is a viable means for integrating heterogeneous knowled...
Under special circumstances, summaries should conform to a particular st...
This paper presents our semantic parsing system for the evaluation task ...
Entity alignment is the task of linking entities with the same real-worl...
We study learning of a matching model for response selection in
retrieva...
We present a document-grounded matching network (DGMN) for response sele...