Summary of Hybrinfox at Checkthat! 2024 — Task 2: Enriching Bert Models with the Expert System Vago For Subjectivity Detection, by Morgane Casanova et al.
HYBRINFOX at CheckThat! 2024 – Task 2: Enriching BERT Models with the Expert System VAGO for Subjectivity Detection
by Morgane Casanova, Julien Chanson, Benjamin Icard, Géraud Faye, Guillaume Gadek, Guillaume Gravier, Paul Égré
First submitted to arxiv on: 4 Jul 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The HYBRINFOX method combines a RoBERTa model fine-tuned for subjectivity detection with a frozen sentence-BERT (sBERT) model to capture semantics. The method uses several scores calculated by the English version of the VAGO expert system to measure vagueness and subjectivity in texts based on the lexicon. In English, HYBRINFOX ranked 1st with a macro F1 score of 0.7442. For other languages, translation into English was used, producing mixed results. The paper outlines ways to improve the method for languages besides English. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research uses a special way to detect how subjective or objective something is in text. It combines two types of AI models: RoBERTa and sentence-BERT. This helps them understand what words mean and how they relate to each other. They also use a system called VAGO to measure how vague or subjective the text is. In English, this method was really good at detecting subjectivity (1st place). For other languages, it didn’t do as well, but that’s okay because they’re still working on improving it. |
Keywords
» Artificial intelligence » Bert » F1 score » Semantics » Translation
