These are some of the retrieval training datasets used for training RaDeR models, sonsisting of different types of query combinations.
RaDeR
non-profit
AI & ML interests
None defined yet.
These are the paths to the Lora weights of our best RaDeR retrievers.
These are the merged retriever checkpoints for the best models from our RaDeR paper.
This contains RaDeR retrievers trained from the Qwen-2.5-instruct family of models of different sizes (3B,7B,14B)
-
Raderspace/RaDeR_Qwen25-14B_NuminaMath_MATH_allquerytypes
Feature Extraction • 14B • Updated • 311 -
Raderspace/RaDeR_Qwen25_3B_NuminaMath_MATH_allquerytypes
Feature Extraction • 3B • Updated • 378 • 2 -
Raderspace/RaDeR_Qwen25-7B_NuminaMath_MATH_allquerytypes
Feature Extraction • 7B • Updated • 311 • 1
These are the best reranker models from our RaDeR paper with LoRA weights.
These are some of the retrieval training datasets used for training RaDeR models, sonsisting of different types of query combinations.
This contains RaDeR retrievers trained from the Qwen-2.5-instruct family of models of different sizes (3B,7B,14B)
-
Raderspace/RaDeR_Qwen25-14B_NuminaMath_MATH_allquerytypes
Feature Extraction • 14B • Updated • 311 -
Raderspace/RaDeR_Qwen25_3B_NuminaMath_MATH_allquerytypes
Feature Extraction • 3B • Updated • 378 • 2 -
Raderspace/RaDeR_Qwen25-7B_NuminaMath_MATH_allquerytypes
Feature Extraction • 7B • Updated • 311 • 1
These are the paths to the Lora weights of our best RaDeR retrievers.
These are the best reranker models from our RaDeR paper with LoRA weights.
These are the merged retriever checkpoints for the best models from our RaDeR paper.