Papers by Benjamin Icard

2 papers
Embedding Style Beyond Topics: Analyzing Dispersion Effects Across Different Language Models (2025.coling-main)

Copied to clipboard

Challenge: Using a literary corpus that alternates between topics and styles, we compare language models across French and English.
Approach: They analyze how writing style affects embedding spaces across multiple language models . they use a literary corpus that alternates between topics and styles to compare their results .
Outcome: The proposed model is based on two established literary works in French and English.
A Multi-Label Dataset of French Fake News: Human and Machine Insights (2024.lrec-main)

Copied to clipboard

Challenge: a corpus of documents selected from 17 sources of french press considered unreliable by experts is annotated using 11 labels by 8 annotators.
Approach: They present a corpus of documents annotated using 11 labels by 8 annotators . they use a subjectivity analyzer VAGO to clarify the link between subjective and fake news labels .
Outcome: The proposed dataset identifies features that humans consider as characteristic of fake news . it also clarifies the link between ascriptions of the label Subjective and ascribeds of Fake News .

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