Data artifacts related to the paper "ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning".
AI & ML interests
Natural language processing group at Columbia University
Models trained using the LION pipeline. Paper: https://arxiv.org/abs/2407.06542; Code: https://github.com/Columbia-NLP-Lab/LionAlignment
-
Columbia-NLP/LION-LLaMA-3-8b-odpo-v1.0
Text Generation • 8B • Updated • 12 • 2 -
Columbia-NLP/LION-LLaMA-3-8b-dpo-v1.0
Text Generation • 8B • Updated • 10 • 2 -
Columbia-NLP/LION-LLaMA-3-8b-sft-v1.0
Text Generation • 8B • Updated • 10 -
Columbia-NLP/LION-Gemma-2b-odpo-v1.0
Text Generation • 3B • Updated • 6 • 4
Data artifacts related to the paper "ExACT: Teaching AI Agents to Explore with Reflective-MCTS and Exploratory Learning".
Datasets used to train the LION pipeline. Paper: https://arxiv.org/abs/2407.06542; Code: https://github.com/Columbia-NLP-Lab/LionAlignment
Models trained using the LION pipeline. Paper: https://arxiv.org/abs/2407.06542; Code: https://github.com/Columbia-NLP-Lab/LionAlignment
-
Columbia-NLP/LION-LLaMA-3-8b-odpo-v1.0
Text Generation • 8B • Updated • 12 • 2 -
Columbia-NLP/LION-LLaMA-3-8b-dpo-v1.0
Text Generation • 8B • Updated • 10 • 2 -
Columbia-NLP/LION-LLaMA-3-8b-sft-v1.0
Text Generation • 8B • Updated • 10 -
Columbia-NLP/LION-Gemma-2b-odpo-v1.0
Text Generation • 3B • Updated • 6 • 4