PA-RAG: RAG Alignment via Multi-Perspective Preference Optimization

πŸŽ‰πŸŽ‰πŸŽ‰ PA-RAG is accepted by NAACL 2025!

Paper Link: https://arxiv.org/pdf/2412.14510
Github Link: https://github.com/wujwyi/PA-RAG

This is a model fine-tuned on Meta-Llama-3-8B-Instruct using PA-RAG. The training data for PA-RAG, available at Huggingface link or Google Drive link

Downloads last month
10
Safetensors
Model size
8.03B params
Tensor type
BF16
Β·
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Model tree for wuqiong1/PA-RAG_Meta-Llama-3-8B-Instruct

Finetuned
(582)
this model
Quantizations
2 models

Dataset used to train wuqiong1/PA-RAG_Meta-Llama-3-8B-Instruct

Collection including wuqiong1/PA-RAG_Meta-Llama-3-8B-Instruct