Pre-trained models have achieved success in Chinese Short Text Matching ...
Consistency regularization methods, such as R-Drop (Liang et al., 2021) ...
Multilingual sentence representations are the foundation for similarity-...
Current Transformer-based natural language understanding (NLU) models he...
Named entity recognition in real-world applications suffers from the
div...
The multilingual neural machine translation (NMT) model has a promising
...
Recent studies reveal that various biases exist in different NLP tasks, ...
In recent years, there has been an increased popularity in image and spe...
Task-agnostic knowledge distillation attempts to address the problem of
...
The challenge of information extraction (IE) lies in the diversity of la...
The quality of knowledge retrieval is crucial in knowledge-intensive
con...
Software engineers working with the same programming language (PL) may s...
Recent cross-lingual cross-modal works attempt to extend Vision-Language...
Speech representation learning has improved both speech understanding an...
Recently, the practical deployment of open-domain dialogue systems has b...
Recent progress in diffusion models has revolutionized the popular techn...
Assigning items to owners is a common problem found in various real-worl...
Derivative-free prompt learning has emerged as a lightweight alternative...
Dialogue contradiction is a critical issue in open-domain dialogue syste...
Existing pipelined task-oriented dialogue systems usually have difficult...
Recent years have witnessed the rise and success of pre-training techniq...
Recent Vision-Language Pre-trained (VLP) models based on dual encoder ha...
Many open-domain dialogue models pre-trained with social media comments ...
As the first session-level Chinese dataset, CHASE contains two separate
...
Molecular property prediction is a fundamental task in the drug and mate...
While pre-trained language models (LMs) have brought great improvements ...
AI-based protein structure prediction pipelines, such as AlphaFold2, hav...
Generative open-domain dialogue systems can benefit from external knowle...
We introduce Bi-SimCut: a simple but effective training strategy to boos...
Many recent works indicate that the deep neural networks tend to take da...
While there is increasing concern about the interpretability of neural
m...
Neural retrievers based on pre-trained language models (PLMs), such as
d...
Accurate ADMET (an abbreviation for "absorption, distribution, metabolis...
Emotional support is a crucial skill for many real-world scenarios, incl...
Most dialog systems posit that users have figured out clear and specific...
Information extraction suffers from its varying targets, heterogeneous
s...
In this paper, we present DuReader_retrieval, a large-scale Chinese data...
Despite recent progress of pre-trained language models on generating flu...
Vision-Language Pre-training (VLP) has achieved impressive performance o...
Due to the limitations of the model structure and pre-training objective...
Most of the open-domain dialogue models tend to perform poorly in the se...
Conventional methods for the image-text generation tasks mainly tackle t...
Pre-trained language models have achieved state-of-the-art results in va...
Task-oriented dialogue systems have been plagued by the difficulties of
...
In this paper, we focus on studying robustness evaluation of Chinese que...
Machine learning shows great potential in virtual screening for drug
dis...
In task-oriented dialogue systems, recent dialogue state tracking method...
Most of existing extractive multi-document summarization (MDS) methods s...
Pre-trained language models (PLMs), such as BERT and GPT, have revolutio...
To explore the limit of dialogue generation pre-training, we present the...