Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userLocality Preserving Sentence Encoding
تحافظ الموقع على ترميز الجملة
191
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
على الرغم من أن الأبحاث حول Word Advedings قد حققت تقدما كبيرا في السنوات الأخيرة، فإن العديد من المهام في معالجة اللغة الطبيعية موجودة على مستوى الجملة. وبالتالي، من الضروري تعلم تضمين الجملة. في الآونة الأخيرة، يقترح عقوبة برت (SBERT) تعلم التشريح على مستوى الجملة، ويستخدم المنتج الداخلي (أو، التشابه الجيبكي) لحساب التشابه الدلالي بين الجمل. ومع ذلك، لا يمكن لهذا القياس وصف الهياكل الدلالية بين الجمل. السبب هو أن الجمل قد تكمن في مشعب في الفضاء المحيط بدلا من توزيع الفضاء الإقليدي. وبالتالي، فإن التشابه التجميل لا يمكن أن تقريب المسافات على المنوع. لمعالجة المشكلة الشديدة، نقترح طريقة تضمين عقوبة جديدة تسمى برت حكم مع الحفاظ على محلية (SBERT-LP)، والتي تكتشف الجملة الفرعية من الفضاء العالي الأبعاد ويمنح مجموعة فرعية تمثيل جملة مدمجة من قبل الهياكل الهندسية المحفوظة محليا من الجمل وبعد قارننا Sbert-LP مع العديد من نهج تضمين الجملة الحالية من ثلاثة وجهات نظر: تشابه الجملة، تصنيف الجملة وتجميع الجملة. تظهر النتائج التجريبية ودراسات الحالة أن طريقتنا ترميز الجمل بشكل أفضل بمعنى الهياكل الدلالية.
Although researches on word embeddings have made great progress in recent years, many tasks in natural language processing are on the sentence level. Thus, it is essential to learn sentence embeddings. Recently, Sentence BERT (SBERT) is proposed to learn embeddings on the sentence level, and it uses the inner product (or, cosine similarity) to compute semantic similarity between sentences. However, this measurement cannot well describe the semantic structures among sentences. The reason is that sentences may lie on a manifold in the ambient space rather than distribute in an Euclidean space. Thus, cosine similarity cannot approximate distances on the manifold. To tackle the severe problem, we propose a novel sentence embedding method called Sentence BERT with Locality Preserving (SBERT-LP), which discovers the sentence submanifold from a high-dimensional space and yields a compact sentence representation subspace by locally preserving geometric structures of sentences. We compare the SBERT-LP with several existing sentence embedding approaches from three perspectives: sentence similarity, sentence classification and sentence clustering. Experimental results and case studies demonstrate that our method encodes sentences better in the sense of semantic structures.
References used
https://aclanthology.org/
rate research
Read More
A private learning scheme TextHide was recently proposed to protect the private text data during the training phase via so-called instance encoding. We propose a novel reconstruction attack to break TextHide by recovering the private training data, a
nd thus unveil the privacy risks of instance encoding. We have experimentally validated the effectiveness of the reconstruction attack with two commonly-used datasets for sentence classification. Our attack would advance the development of privacy preserving machine learning in the context of natural language processing.
Automatic news recommendation has gained much attention from the academic community and industry. Recent studies reveal that the key to this task lies within the effective representation learning of both news and users. Existing works typically encod
e news title and content separately while neglecting their semantic interaction, which is inadequate for news text comprehension. Besides, previous models encode user browsing history without leveraging the structural correlation of user browsed news to reflect user interests explicitly. In this work, we propose a news recommendation framework consisting of collaborative news encoding (CNE) and structural user encoding (SUE) to enhance news and user representation learning. CNE equipped with bidirectional LSTMs encodes news title and content collaboratively with cross-selection and cross-attention modules to learn semantic-interactive news representations. SUE utilizes graph convolutional networks to extract cluster-structural features of user history, followed by intra-cluster and inter-cluster attention modules to learn hierarchical user interest representations. Experiment results on the MIND dataset validate the effectiveness of our model to improve the performance of news recommendation.
Transformer models are permutation equivariant. To supply the order and type information of the input tokens, position and segment embeddings are usually added to the input. Recent works proposed variations of positional encodings with relative posit
ion encodings achieving better performance. Our analysis shows that the gain actually comes from moving positional information to attention layer from the input. Motivated by this, we introduce Decoupled Positional Attention for Transformers (DIET), a simple yet effective mechanism to encode position and segment information into the Transformer models. The proposed method has faster training and inference time, while achieving competitive performance on GLUE, XTREME and WMT benchmarks. We further generalize our method to long-range transformers and show performance gain.
This paper describes N-XKT (Neural encoding based on eXplanatory Knowledge Transfer), a novel method for the automatic transfer of explanatory knowledge through neural encoding mechanisms. We demonstrate that N-XKT is able to improve accuracy and gen
eralization on science Question Answering (QA). Specifically, by leveraging facts from background explanatory knowledge corpora, the N-XKT model shows a clear improvement on zero-shot QA. Furthermore, we show that N-XKT can be fine-tuned on a target QA dataset, enabling faster convergence and more accurate results. A systematic analysis is conducted to quantitatively analyze the performance of the N-XKT model and the impact of different categories of knowledge on the zero-shot generalization task.
It has been widely recognized that syntax information can help end-to-end neural machine translation (NMT) systems to achieve better translation. In order to integrate dependency information into Transformer based NMT, existing approaches either expl
oit words' local head-dependent relations, ignoring their non-local neighbors carrying important context; or approximate two words' syntactic relation by their relative distance on the dependency tree, sacrificing exactness. To address these issues, we propose global positional encoding for dependency tree, a new scheme that facilitates syntactic relation modeling between any two words with keeping exactness and without immediate neighbor constraint. Experiment results on NC11 German→English, English→German and WMT English→German datasets show that our approach is more effective than the above two strategies. In addition, our experiments quantitatively show that compared with higher layers, lower layers of the model are more proper places to incorporate syntax information in terms of each layer's preference to the syntactic pattern and the final performance.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
