Generative AI Weekly Research Highlights | Mar'24 Part 1

Disclaimer: The content in this video is AI-generated and adheres to YouTube's guidelines. Each video undergoes manual review and curation before publishing to ensure accuracy and quality.

Summary of Research Papers:

1. On Protecting the Data Privacy of Large Language Models (LLMs): A Survey [https://arxiv.org/pdf/2403.05156.pdf]

2. Automatic and Universal Prompt Injection Attacks against Large Language Models [https://arxiv.org/pdf/2403.04957.pdf]

3. SheetAgent: A Generalist Agent for Spreadsheet Reasoning and Manipulation via Large Language Models [https://arxiv.org/pdf/2403.03636.pdf]

4. Angry Men, Sad Women: Large Language Models Reflect Gendered Stereotypes in Emotion Attribution [https://arxiv.org/pdf/2403.03121.pdf]

5. OFFLANDAT: A COMMUNITY BASED IMPLICIT OFFENSIVE LANGUAGE DATASET GENERATED BY LARGE LANGUAGE MODEL THROUGH PROMPT ENGINEERING [https://arxiv.org/pdf/2403.02472.pdf]

6. On Prompt-Driven Safeguarding for Large Language Models [https://arxiv.org/pdf/2401.18018.pdf]

7. Benchmarking and Defending Against Indirect Prompt Injection Attacks on Large Language Models [https://arxiv.org/pdf/2312.14197.pdf]

8. Adapting Large Language Models for Content Moderation: Pitfalls in Data Engineering and Supervised Fine-tuning [https://arxiv.org/pdf/2310.03400.pdf]

00:00 Intro

00:20 Data Privacy in LLMs

00:47 Prompt Injection Attacks

01:07 SheetAgent for Spreadsheets

01:27 Gender Stereotypes and Emotions

01:48 Offensive Language Detection

02:08 Safeguarding LLMs

02:30 Defending Against Indirect Prompt Injection Attacks

02:50 Content Moderation via LLMs

03:17 End

#generativeai,#promptengineering,#largelanguagemodels,#openai,#chatgpt,#gpt4,#ai,#abcp,#prompt,#responsibleai,