Summary of Leveraging Large Language Models For Suicide Detection on Social Media with Limited Labels, by Vy Nguyen et al.
Leveraging Large Language Models for Suicide Detection on Social Media with Limited Labels
by Vy Nguyen, Chau Pham
First submitted to arxiv on: 6 Oct 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
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 paper explores the use of Large Language Models (LLMs) to automatically detect suicidal content in text-based social media posts. The authors propose a novel method for generating pseudo-labels for unlabeled data by prompting LLMs, and fine-tuning traditional classification techniques to enhance label accuracy. They develop an ensemble approach using multiple models such as Qwen2-72B-Instruct, Llama3-8B, Llama3.1-8B, Gemma2-9B, and evaluate their approach on the Suicide Ideation Detection on Social Media Challenge dataset. The results show that the ensemble model improves detection accuracy by 5% points compared to individual models, achieving a weight F1 score of 0.770 on the public test set and 0.731 on the private test set. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper uses Large Language Models (LLMs) to find suicidal thoughts in social media posts. This helps people who are struggling to get help early. The authors use special techniques to make the models better at detecting these posts. They tested their approach using a big dataset and found that combining multiple models worked really well, making it easier to identify when someone is having suicidal thoughts. |
Keywords
» Artificial intelligence » Classification » Ensemble model » F1 score » Fine tuning » Prompting
