Papers by Hanbin Ko

2 papers
CoNAN: A Complementary Neighboring-based Attention Network for Referring Expression Generation (2020.coling-main)

Copied to clipboard

Challenge: Existing approaches to identify and describe complex daily scenes are limited by ambiguity.
Approach: They propose a Complementary Neighboring-based Attention Network that utilizes visual differences between the target object and its highly-related neighbors as complementary features.
Outcome: The proposed expression outperforms state-of-the-art models on a dataset of 3D objects.
CREPE: Rapid Chest X-ray Report Evaluation by Predicting Multi-category Error Counts (2025.emnlp-main)

Copied to clipboard

Challenge: CREPE is a metric for rapid, interpretable, and clinically grounded automated chest X-ray report generation.
Approach: They propose to use a domain-specific BERT model fine-tuned with a multi-head regression architecture to predict multi-category error counts across six clinically meaningful categories.
Outcome: CREPE outperforms traditional and recent metrics on a large-scale synthetic dataset of 32,000 annotated report pairs.

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