Papers with systematic model

2 papers
Design Challenges and Misconceptions in Neural Sequence Labeling (C18-1)

Copied to clipboard

Challenge: Existing neural sequence labeling models have been used for many tasks such as POS tagging, chunking and named entity recognition (NER).
Approach: They propose to replicate twelve neural sequence labeling models and compare them to three benchmarks to find out which models are effective and which are inconsistent.
Outcome: The proposed models are compared on NER, Chunking, and POS tagging benchmarks.
Explaining Matters: Leveraging Definitions and Semantic Expansion for Sexism Detection (2025.acl-long)

Copied to clipboard

Challenge: Existing tools for sexism detection fail to capture subtle distinctions within sexist content, limiting their practical applicability.
Approach: They propose two techniques to address class imbalance and nuanced nature of sexist language . definition-based data augmentation leverages category-specific definitions to generate semantically-aligned examples .
Outcome: The proposed techniques improve accuracy across all tasks and improve reliability.

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