Due to the unbalanced training data distribution, the language ability o...
Audio-guided Video Object Segmentation (A-VOS) and Referring Video Objec...
Vision-language models (VLMs) have shown impressive performance in
subst...
Generating visually grounded image captions with specific linguistic sty...
Automatic metrics play a crucial role in machine translation. Despite th...
Neural machine translation has achieved promising results on many transl...
Large-scale Pretrained Language Models (LLMs), such as ChatGPT and GPT4,...
Interactive Image Segmentation (IIS) has emerged as a promising techniqu...
Large language models (LLMs) have demonstrated remarkable potential in
h...
Augmenting the base neural model with a token-level symbolic datastore i...
Abstractive summarization is the process of generating a summary given a...
kNN-MT presents a new paradigm for domain adaptation by building an exte...
As one of the challenging NLP tasks, designing math word problem (MWP)
s...
In recent years, vision and language pre-training (VLP) models have adva...
In this paper, we consider a novel research problem, music-to-text
synae...
3D reconstruction of pulmonary segments plays an important role in surgi...
The numerical reasoning in the financial domain – performing quantitativ...
Complaining is a speech act that expresses a negative inconsistency betw...
Recently, parallel text generation has received widespread attention due...
We study the problem of online learning with human feedback in the
human...
Recently, kNN-MT has shown the promising capability of directly
incorpor...
Unknown intent detection aims to identify the out-of-distribution (OOD)
...
Link prediction in large-scale knowledge graphs has gained increasing
at...
kNN-MT, recently proposed by Khandelwal et al. (2020a), successfully com...
Machine Translation Quality Estimation (QE) is a task of predicting the
...
Sequence-to-sequence (seq2seq) problems such as machine translation are
...
Non-autoregressive Transformer is a promising text generation model. How...
It is good practice to name test methods such that they are comprehensib...
Social recommendation is effective in improving the recommendation
perfo...
Previous domain adaptation research usually neglect the diversity in
tra...
Aspect-based sentiment analysis (ABSA) aims at analyzing the sentiment o...
Aspect-level sentiment classification (ALSC) and aspect oriented opinion...
Discourse context has been proven useful when translating documents. It ...
Predicting clinical outcome is remarkably important but challenging. Res...
Recent studies show that the attention heads in Transformer are not equa...
Cross-prompt automated essay scoring (AES) requires the system to use no...
We propose a novel mechanism to improve a text generator with a
discrimi...
We propose a novel mechanism to improve a text generator with a
discrimi...
Transformer, based on the encoder-decoder framework, has achieved
state-...
Document-level machine translation manages to outperform sentence level
...
Target-oriented opinion words extraction (TOWE) is a new subtask of ABSA...
Non-autoregressive models are promising on various text generation tasks...
Transformer model has been widely used on machine translation tasks and
...
Neural machine translation systems tend to fail on less de-cent inputs
d...
Natural Language Inference (NLI) aims to determine the logic relationshi...
Graph neural networks (GNNs) have shown great power in learning on attri...
In sequence labeling, previous domain adaptation methods focus on the
ad...
Monolingual data has been demonstrated to be helpful in improving the
tr...
Relation detection is a core step in many natural language process
appli...
State-of-the-art machine translation models are still not on par with hu...