abel
abel copied to clipboard
GAIR-NLP
SOTA Math Opensource LLM
Hi, Could you please add a new baseline——MuggleMATH in the Leaderboard for Mathematical Reasoning on the github webpage? MuggleMATH mainly investigates the scaling law and generalization for data augmentation for...
I mean tell the readers what have you done in details. I'm sorry but I cannot get enough information from this repo for now (2023/09/21 14:28 UTC+8).
Relevant Papers: * https://arxiv.org/abs/2305.20050 * https://arxiv.org/abs/2306.01693
We definitely need to continue pretraining our base model on a large number of math-related corpus.
The current model's (Abel) training data and base model constraints limit its ability to provide responses in languages other than English.
we anticipate that the mathematical reasoning abilities enabled by large models can be integrated into chatbots for various domains such as medicine, law, physics, chemistry, etc. The key to achieving...
Currently, the model is limited in supporting diverse question types (arithmetic, proof) and scenarios; it would be nice if we could generalize the model to them.
It would be nice if we could conduct more extensive robust analyses in terms of the following perspectives: * re-evaluate model on adversarial evaluation set * evaluate the faithfulness of...