Search Results for author:
Found 5 papers, 0 papers with code
Knowledge Unlearning for LLMs: Tasks, Methods, and Challenges
We further present evaluation datasets used in existing methods, and finally conclude this survey by presenting the ongoing challenges and future directions.
Tuning Large language model for End-to-end Speech Translation
The training of LST consists of two stages: (1) Modality adjustment, where the adapter is tuned to align speech representation with text embedding space, and (2) Downstream task fine-tuning, where both the adapter and LLM model are trained to optimize performance on the E2EST task.
DropDim: A Regularization Method for Transformer Networks
In contrast to the general dropout method, which randomly drops neurons, DropDim drops part of the embedding dimensions.
Improving Speech Translation by Cross-Modal Multi-Grained Contrastive Learning
However, the final model often performs worse on the MT task than the MT model trained alone, which means that the knowledge transfer ability of this method is also limited.
Decouple Non-parametric Knowledge Distillation For End-to-end Speech Translation
Existing techniques often attempt to make knowledge transfer from a powerful machine translation (MT) to speech translation (ST) model with some elaborate techniques, which often requires transcription as extra input during training.
