I would be interested in a model trained to take in N parameters, and output B outputs. I would use it for robotics such that I can input all of the state of the motors, sensors, camera, etc. and output actions the robot could take. Could be used for practically anything and would streamline the use of LLMs by converging them to a single inference step.
Related topics
Topic | Replies | Views | Activity | |
---|---|---|---|---|
Multi-LLMs + API Assistant System | 0 | 454 | November 24, 2023 | |
N chat completion choices with follow-up replies | 6 | 1691 | August 18, 2023 | |
Multimodal queries with voice | 0 | 1066 | May 15, 2024 | |
Fine tuning a model with multiple parameters | 0 | 91 | November 22, 2024 | |
Feature Request: First-class Object References for Function Calls | 3 | 789 | December 2, 2023 |