Layer Increasing Network Scaling (LiNeS) - Preventing Catastrophic Forgetting in Language Models

Layer Increasing Network Scaling (LiNeS) - Preventing Catastrophic Forgetting in Language Models

Discover AI via YouTube Direct link

Task vectors are not task vectors

9 of 9

9 of 9

Task vectors are not task vectors

Class Central Classrooms beta

YouTube videos curated by Class Central.

Classroom Contents

Layer Increasing Network Scaling (LiNeS) - Preventing Catastrophic Forgetting in Language Models

Automatically move to the next video in the Classroom when playback concludes

  1. 1 VLMs suffer from catastrophic forgetting
  2. 2 ChatGPT SEARCH 4omni
  3. 3 ChatGPT SEARCH hallucinates complete study
  4. 4 Post-training Layer Scaling VLM, LLM
  5. 5 Simple visualization of LiNeS method
  6. 6 Multi-task to multi-model merge
  7. 7 AdaMerging multi-task learning
  8. 8 Anisotropic scaling
  9. 9 Task vectors are not task vectors

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.