In this paper, we introduce a new task for code completion that focuses ...
Large Language Models (LLMs) have greatly advanced code auto-completion
...
Chat models, such as ChatGPT, have shown impressive capabilities and hav...
We present Mirror, an open-source platform for data exploration and anal...
Masked language modeling is widely used for pretraining large language m...
Intermediate-task transfer can benefit a wide range of NLP tasks with
pr...
Prompt-based learning (i.e., prompting) is an emerging paradigm for
expl...
In this paper, we propose LaPraDoR, a pretrained dual-tower dense retrie...
Language models (LMs) can reproduce (or amplify) toxic language seen dur...
With recent developments in new architectures like Transformer and
pretr...
Effectively scaling large Transformer models is a main driver of recent
...
PromptSource is a system for creating, sharing, and using natural langua...
Large language models have recently been shown to attain reasonable zero...
Recent studies on compression of pretrained language models (e.g., BERT)...
The scale, variety, and quantity of publicly-available NLP datasets has ...
We present Meta Learning for Knowledge Distillation (MetaDistil), a simp...
Cant is important for understanding advertising, comedies and dog-whistl...
In this paper, we propose Patience-based Early Exit, a straightforward y...
Recently, large-scale datasets have vastly facilitated the development i...
In this paper, we propose a novel model compression approach to effectiv...
Current neural Natural Language Generation (NLG) models cannot handle
em...
Identifying the named entities mentioned in text would enrich many seman...
Recently, with the prevalence of large-scale image dataset, the co-occur...
In recent years, with the prevalence of social media and smart devices,
...