Skip to search formSkip to main contentSkip to account menu

Index-ASR Technical Report

The core idea of Index-ASR lies in the integration of LLM and large-scale training data enriched with background noise and contextual information, which achieves strong performance on both open-source benchmarks and in-house test sets, highlighting its robustness and practicality for real-world ASR applications.
arXiv (opens in a new tab)

CoSPED: Consistent Soft Prompt Targeted Data Extraction and Defense

This work proposes CoSPED, short for Consistent Soft Prompt targeted data Extraction and Defense, and introduces several innovative components, including Dynamic Loss, Additive Loss, Common Loss, and Self Consistency Decoding Strategy, and tested to enhance the consistency of the soft prompt tuning process.
arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)