Papers with sentiment representation

2 papers
SKEP: Sentiment Knowledge Enhanced Pre-training for Sentiment Analysis (2020.acl-main)

Copied to clipboard

Challenge: sentiment knowledge is ignored in sentiment analysis, despite its use in pretraining.
Approach: They propose to use sentiment knowledge to learn a unified sentiment representation for multiple sentiment analysis tasks.
Outcome: The proposed method outperforms strong pre-training baseline on three kinds of sentiment tasks.
Improving Personalized Sentiment Representation with Knowledge-enhanced and Parameter-efficient Layer Normalization (2024.lrec-main)

Copied to clipboard

Challenge: Existing studies on personalized sentiment classification consider document reviews as overall text unit and incorporate backgrounds (i.e., user and product information) Existing methods for personalized sentiment modeling have quadratic costs that increase with text length and heterogeneous mixes of background information and textual information.
Approach: They propose a knowledge-enhanced and parameter-efficient layer normalization model that leverages pretrained checkpoints and background information into transformer structures.
Outcome: The proposed model can be used to improve pretrained language models in document reviews and incorporate background information with parameter-efficient fine-tuning and knowledge injecting.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations