A unified end-to-end retriever-reader framework for knowledge-based vqa
Knowledge-based Visual Question Answering (VQA) expects models to rely on external
knowledge for robust answer prediction. Though significant it is, this paper discovers several …
knowledge for robust answer prediction. Though significant it is, this paper discovers several …
MERIt: Meta-path guided contrastive learning for logical reasoning
Logical reasoning is of vital importance to natural language understanding. Previous studies
either employ graph-based models to incorporate prior knowledge about logical relations, or …
either employ graph-based models to incorporate prior knowledge about logical relations, or …
Machine reading comprehension using case-based reasoning
We present an accurate and interpretable method for answer extraction in machine reading
comprehension that is reminiscent of case-based reasoning (CBR) from classical AI. Our …
comprehension that is reminiscent of case-based reasoning (CBR) from classical AI. Our …
Enhanced Multi-Domain Dialogue State Tracker With Second-Order Slot Interactions
Dialogue state tracking (DST) is often used to track the system's understanding of the user
goal in task-oriented dialogue systems. Existing DST methods mainly fall into two categories …
goal in task-oriented dialogue systems. Existing DST methods mainly fall into two categories …
Bridging the gap between language models and cross-lingual sequence labeling
Large-scale cross-lingual pre-trained language models (xPLMs) have shown effectiveness
in cross-lingual sequence labeling tasks (xSL), such as cross-lingual machine reading …
in cross-lingual sequence labeling tasks (xSL), such as cross-lingual machine reading …
DC-Graph: a chunk optimization model based on document classification and graph learning
J Zhou, G Zhang, O Alfarraj, A Tolba, X Li… - Artificial Intelligence …, 2024 - Springer
Existing machine reading comprehension methods use a fixed stride to chunk long texts,
which leads to missing contextual information at the boundaries of the chunks and a lack of …
which leads to missing contextual information at the boundaries of the chunks and a lack of …
Leveraging greater relations for improving multi-choice reading comprehension
H Yan, L Liu, X Feng, Q Huang - Neural Computing and Applications, 2022 - Springer
Remarkable success has been achieved in the last few years on machine reading
comprehension tasks. In previous works, long-range dependencies were captured by …
comprehension tasks. In previous works, long-range dependencies were captured by …
RoBERTa-CoA: RoBERTa-Based Effective Finetuning Method Using Co-Attention
In the field of natural language processing, artificial intelligence (AI) technology has been
utilized to solve various problems, such as text classification, similarity measurement …
utilized to solve various problems, such as text classification, similarity measurement …
基于多级语义对齐的图像-文本匹配算法
李艺茹， 姚涛， 张林梁， 孙玉娟， 付海燕 - 北京航空航天大学学报, 2022 - bhxb.buaa.edu.cn
图像中的区域特征更关注于图像中的前景信息, 背景信息往往被忽略, 如何有效的联合局部特征
和全局特征还没有得到充分地研究. 为解决上述问题, 加强全局概念和局部概念之间的关联得到 …
和全局特征还没有得到充分地研究. 为解决上述问题, 加强全局概念和局部概念之间的关联得到 …
Exploring Self-supervised Logic-enhanced Training for Large Language Models
Existing efforts to improve logical reasoning ability of language models have predominantly
relied on supervised fine-tuning, hindering generalization to new domains and/or tasks. The …
relied on supervised fine-tuning, hindering generalization to new domains and/or tasks. The …