This paper develops a Decentralized Multi-Agent Reinforcement Learning
(...
Existing large language models (LLMs) can only afford fix-sized inputs d...
Scientific literature understanding tasks have gained significant attent...
This paper explores the effectiveness of model-generated signals in impr...
The retrieval model is an indispensable component for real-world
knowled...
Backpropagation, the cornerstone of deep learning, is limited to computi...
Learning transferable representation of knowledge graphs (KGs) is challe...
Radars are widely used to obtain echo information for effective predicti...
Inductive reasoning is a core component of human intelligence. In the pa...
Transformer models have achieved superior performance in various natural...
Standard fine-tuning of large pre-trained language models (PLMs) for
dow...
We propose a novel open-domain question answering (ODQA) framework for
a...
Neural architecture search (NAS) has demonstrated promising results on
i...
Given its effectiveness on knowledge-intensive natural language processi...
Recent years have witnessed a trend of applying context frames to boost ...
Deep generative models (DGMs) are data-eager. Essentially, it is because...
Fine-tuning large-scale pre-trained language models to downstream tasks
...
Human language is grounded on multimodal knowledge including visual know...
We present an efficient method of pretraining large-scale autoencoding
l...
Hyperparameter (HP) tuning in deep learning is an expensive process,
pro...
With the increasing of model capacity brought by pre-trained language mo...
Recent research has shown the existence of significant redundancy in lar...
The bound of the information transmission rate of direct current biased
...
Knowledge distillation (KD) methods compress large models into smaller
s...
Entity linking faces significant challenges, such as prolific variations...
Motivation: A perennial challenge for biomedical researchers and clinica...
Most of today's AI systems focus on using self-attention mechanisms and
...
Most recent progress in natural language understanding (NLU) has been dr...
Due to its potential for a universal interface over both data and text,
...
We present a new method LiST for efficient fine-tuning of large pre-trai...
Sparsely activated models (SAMs), such as Mixture-of-Experts (MoE), can
...
Background: Type-4 clones refer to a pair of code snippets with similar
...
Adversarial regularization can improve model generalization in many natu...
We consider the inverse source problems with multi-frequency sparse near...
This paper addresses robust beamforming design for rate splitting multip...
This paper investigates the inverse scattering problems using sampling
m...
We introduce two data completion algorithms for the limited-aperture pro...
The Lottery Ticket Hypothesis suggests that an over-parametrized network...
We present a simple yet effective Targeted Adversarial Training (TAT)
al...
Adversarial training has been shown to improve the generalization perfor...
Surveillance cameras are widely applied for indoor occupancy measurement...
Misconfigurations have become the dominant causes of software failures i...
Existing curriculum learning approaches to Neural Machine Translation (N...
Air pollution has altered the Earth radiation balance, disturbed the
eco...
We present a new high-order accurate Lagrangian discontinuous Galerkin (...
Applications depend on libraries to avoid reinventing the wheel. Librari...
Current open-domain question answering (QA) systems often follow a
Retri...
To date, most of recent work under the retrieval-reader framework for
op...
We review the EfficientQA competition from NeurIPS 2020. The competition...
We address the problem of enhancing model robustness through regularizat...