Timeline for For specific tasks, is it better to fine-tune models on examples or just use prompting with the context of the task?
Current License: CC BY-SA 4.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jan 13, 2023 at 15:09 | comment | added | Imran Q | Thanks! That link is fascinating. I think they use a fixed vector, but human engineered prompts can use a variety of sizes so it might be worth trying out varying embedding sizes | |
| Jan 4, 2023 at 22:17 | comment | added | Rexcirus | ai.googleblog.com/2022/02/… | |
| Jan 4, 2023 at 22:11 | comment | added | Rexcirus | That is what "prompt tuning" and similar techniques do. Training the network + prompt to be accurate, but keeping the network frozen. | |
| Jan 4, 2023 at 21:21 | comment | added | Imran Q | Thanks, i guess one question is how to find the optimal prompt that will approximate fine tuning | |
| Jan 4, 2023 at 21:20 | vote | accept | Imran Q | ||
| Jan 3, 2023 at 20:18 | history | answered | Rexcirus | CC BY-SA 4.0 |