Skip to content

Release LM-Cocktail

Compare
Choose a tag to compare
@staoxiao staoxiao released this 24 Nov 09:22
· 984 commits to master since this release

LM-Cocktail

Merge language models (e.g., Llama, bge) to improve the general ability of models.
This method can be used to:

  • Mitigate the Problem of Catastrophic Forgetting
  • Improve the performance of new tasks without fine-tuning
  • Approximate multitask learning or model ensemble

More details please refer to paper and code