Search

Extracting the relations of two entities on the sentence-level has drawn increasing attention in recent years but remains facing great challenges on document-level, due to the inherent difficulty in recognizing the relations of two entities across multiple sentences. Previous works show that employing the graph convolutional neural network can help the model capture unstructured dependent information of entities. However, they usually employed the non-adaptive weight edges to build the correlation weight matrix which suffered from the problem of information redundancy and gradient disappearance. To solve this problem, we propose a deep gated graph reasoning model for document-level relation extraction, namely, BERT-GGNNs, which employ an improved gated graph neural network with a learnable correlation weight matrix to establish multiple deep gated graph reason layers. The proposed deep gated graph reasoning layers make the model easier to reasoning the relations between entities hidden in the document. Experiments show that the proposed model outperforms most of strong baseline models, and our proposed model is 0.3% and 0.3% higher than the famous LSR-BERT model on the F1 and Ing F1, respectively. [ABSTRACT FROM AUTHOR]

Ultra‐high field (UHF) magnetic resonance imaging (MRI) has emerged as a focal point of interest in the field of cancer diagnosis. Despite the ability of current paramagnetic or superparamagnetic smart MRI contrast agents to selectively enhance tumor signals in low‐field MRI, their effectiveness at UHF remains inadequate due to inherent magnetism. Here, we report a ligand‐mediated magnetism‐conversion nanoprobe (MCNP) composed of 3‐mercaptopropionic acid ligand‐coated silver‐gadolinium bimetallic nanoparticles. The MCNP exhibits a pH‐dependent magnetism conversion from ferromagnetism to diamagnetism, facilitating tunable nanomagnetism for pH‐activatable UHF MRI. Under neutral pH, the thiolate (−S−) ligands lead to short τ′m and increased magnetization of the MCNPs. Conversely, in the acidic tumor microenvironment, the thiolate ligands are protonated and transform into thiol (−SH) ligands, resulting in prolonged τ′m and decreased magnetization of the MCNP, thereby enhancing longitudinal relaxivity (r1) values at UHF MRI. Notably, under a 9 T MRI field, the pH‐sensitive changes in Ag−S binding affinity of the MCNP lead to a remarkable (>10‐fold) r1 increase in an acidic medium (pH 5.0). In vivo studies demonstrate the capability of MCNPs to amplify MRI signal of hepatic tumors, suggesting their potential as a next‐generation UHF‐tailored smart MRI contrast agent. [ABSTRACT FROM AUTHOR]

Sign language is the main communication way for deaf and hard-of-hearing (i.e., DHH) people, which is unfamiliar to most non-deaf and hard-of-hearing (non-DHH) people. To break down the communication barriers between DHH and non-DHH people and to better promote communication among DHH individuals, we have summarized the research progress on sign language translation. We provide the necessary background on sign language translation and introduce its four subtasks (i.e., sign2gloss2text, sign2text, sign2(gloss+text), and gloss2text). We distill the basic mode of sign language translation (SLT) and introduce the transformer-based framework of SLT. We analyze the main challenges of SLT and propose possible directions for its development. [ABSTRACT FROM AUTHOR]