Intelligence
Hate Speech Detection Using Multi-LLM Architecture
Classification using an ensemble of fine-tuned LLMs with in-context learning.
Notes
Intelligence
- Two-tiered multi-LLM architecture using LLaMA models (1B, 3B)
- 18% faster processing and 54% lower memory usage
- Fine-tuning with Torchtune and ICL
- Proposed: model calibration, LoRA/QLoRA for efficient fine-tuning