Papers by Baoyu Fan

2 papers
Multiplex Graph Neural Network for Extractive Text Summarization (2021.emnlp-main)

Copied to clipboard

Challenge: Existing methods for extractive text summarization do not consider multiple types of inter-sentential relationships, nor model intra-sententential relationships.
Approach: They propose a novel method to combine different types of relationships among sentences and words to model sentence embedding.
Outcome: The proposed model is compared with existing methods on CNN/DailyMail benchmark dataset to demonstrate its effectiveness.
Beyond Scaling: Measuring and Predicting the Upper Bound of Knowledge Retention in Language Model Pre-Training (2026.acl-long)

Copied to clipboard

Challenge: Existing methods to predict performance of large language models are lacking . authors propose a size-dependent mutual information predictor for closed-book question answering accuracy .
Approach: They propose a size-dependent mutual information predictor that integrates knowledge frequency, knowledge specificity, and model size to forecast closed-book question answering accuracy.
Outcome: The proposed method outperforms baseline models and achieves R2 > 0.7 in predicting QA accuracy without additional training.

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