Method, apparatus, device and readable medium for data compression and decompression

A method for data decompression, comprising:

• obtaining a compressed representation of data, wherein the compressed representation is a vectorized representation of the data generated by a first model configured to perform a data compression task on the data, and wherein the first model is constructed based on a language model;
• generating a second input sequence for a second model based on a second prompt and the compressed representation, wherein the second model is constructed based on the language model and is jointly trained with the first model, wherein the second prompt indicates a data compression task to be performed by the second model;
• obtaining a second output sequence of the second model by providing the second input sequence to the second model;
• and
• determining, from the second output sequence, the decompressed data.

An electronic device, comprising:

• at least one processing unit;
• and
• at least one memory coupled to the at least one processing unit and storing instructions executable by the at least one processing unit, the instructions, when executed by the at least one processing unit, causing the electronic device to perform the method acts comprising:
• generating a first input sequence for a first model based on a first prompt and data to be compressed, wherein the first model being is constructed based on a language model, and wherein the first prompt indicates a data compression task to be performed by the first model;
• obtaining a first output sequence of the first model by providing the first input sequence to the first model;
• extracting a compressed representation of the data from the first output sequence, the compressed representation being a vectorized representation of the data;
• and
• decompressing the compressed representation of the data by a second model that is constructed based on the language model and that is jointly trained with the first model.