Domain adaptation of AI NLP encoders with knowledge distillation
Systems, methods, devices, instructions, and other examples are described for natural language processing. One example includes accessing natural language processing general encoder data, where the encoder data is generated from a general-domain dataset that is not domain specific. A domain specific dataset is accessed and filtered encoder data using a subset of the encoder data is generated. The filtered encoder data is trained using the domain specific dataset to generate distilled encoder data, and tuning values for the distilled encoder data are generated to configure task outputs associated with the domain specific dataset.
1. A computer-implemented method comprising:
accessing distilled encoder data, wherein the distilled encoder data is generated by filtering encoder data generated from a general-domain dataset that is not domain specific to generate filtered encoder data;
accessing a stream of domain specific data; and
processing the stream of domain specific data using the distilled encoder data to generate task output data.
2. The computer-implemented method of claim 1 , wherein the stream of domain specific data is domain specific search query data.
3. The computer-implemented method of claim 1 , wherein the stream of domain specific data is real-time chat data from a two-way communication between a user device and an agent device.
4. The computer-implemented method of claim 1 , wherein the domain specific data includes unlabeled data.
5. The computer-implemented method of claim 1 , wherein the distilled encoder data is configured to identify real-time data in a two-way communication associated with an actionable request.
6. The computer-implemented method of claim 1 , wherein processing the stream of domain specific data using the distilled encoder data to generate task output data includes:
submitting strings of data from the stream of domain specific data to a system; and
providing an output from the system to a user or additional artificial intelligence systems for further refining or processing.
7. The computer-implemented method of claim 1 , wherein the task output data includes responses to a user query.
8. A system, comprising:
a memory; and
one or more processors coupled to the memory and configured to perform operations comprising:
accessing distilled encoder data, wherein the distilled encoder data is generated by filtering encoder data generated from a general-domain dataset that is not domain specific to generate filtered encoder data;
accessing a stream of domain specific data; and
processing the stream of domain specific data using the distilled encoder data to generate task output data.
9. The system of claim 8 , wherein the stream of domain specific data is domain specific search query data.
10. The system of claim 8 , wherein the stream of domain specific data is real-time chat data from a two-way communication between a user device and an agent device.
11. The system of claim 8 , wherein the domain specific data includes unlabeled data.
12. The system of claim 8 , wherein the distilled encoder data is configured to identify real-time data in a two-way communication associated with an actionable request.
13. The system of claim 8 , wherein processing the stream of domain specific data using the distilled encoder data to generate task output data includes:
submitting strings of data from the stream of domain specific data to a system; and
providing an output from the system to a user or additional artificial intelligence systems for further refining or processing.
14. The system of claim 8 , wherein the task output data includes responses to a user query.
15. A non-transitory computer readable storage medium comprising instructions that, when executed by one or more processors of a device, cause the device to perform operations comprising:
accessing distilled encoder data, wherein the distilled encoder data is generated by filtering encoder data generated from a general-domain dataset that is not domain specific to generate filtered encoder data;
accessing a stream of domain specific data; and
processing the stream of domain specific data using the distilled encoder data to generate task output data.
16. The non-transitory computer readable storage medium of claim 15 , wherein the stream of domain specific data is domain specific search query data.
17. The non-transitory computer readable storage medium of claim 15 , wherein the stream of domain specific data is real-time chat data from a two-way communication between a user device and an agent device.
18. The non-transitory computer readable storage medium of claim 15 , wherein the domain specific data includes unlabeled data.
19. The non-transitory computer readable storage medium of claim 15 , wherein the distilled encoder data is configured to identify real-time data in a two-way communication associated with an actionable request.
20. The non-transitory computer readable storage medium of claim 15 , wherein processing the stream of domain specific data using the distilled encoder data to generate task output data includes:
submitting strings of data from the stream of domain specific data to a system; and
providing an output from the system to a user or additional artificial intelligence systems for further refining or processing.
21. The non-transitory computer readable storage medium of claim 15 , wherein the task output data includes responses to a user query.