HIGHLIGHTS
- who: Zixi Jia from the Shaanxi Normal University, China have published the research: Vital information matching in vision-and-language navigation, in the Journal: (JOURNAL)
- what: The authors propose a novel model of the key information matching network, VIM-Net, which employs multi-modal matching methods and navigation selftuning modules. If the correct direction is not selected at the beginning, the authors provide a self-tuning module so that the agent can return to the previous position to retrain parameters and then make a new choice. The model proposed by the author still ignores the . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.