43
Facebook Pushes Its Llama 4 AI Model to the Right, Wants to Present “Both Sides” [404 Media] (www.404media.co)
submitted 1 year ago by to c/techtakes@awful.systems
9 comments fedilink hide all child comments
you are viewing a single comment's thread
view the rest of the comments
[–] -2 points 1 year ago (2 children)

Do you have any sources on this? I started looking around for pre-training, training and post-training impact of new input but didn't find what I was looking for. In just my own experience with retraining (e.g. fine-tuning) pre-trained models, it seems to be pretty easy to add or remove data to get significantly different results than the original model.

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 4 child comments
    •
  • [–] 5 points 1 year ago

    It's well-known folklore that reinforcement learning with human feedback (RLHF), the standard post-training paradigm, reduces "alignment," the degree to which a pre-trained model has learned features of reality as it actually exists. Quoting from the abstract of the 2024 paper, Mitigating the Alignment Tax of RLHF (alternate link):

    LLMs acquire a wide range of abilities during pre-training, but aligning LLMs under Reinforcement Learning with Human Feedback (RLHF) can lead to forgetting pretrained abilities, which is also known as the alignment tax.

  • permalink
  • fedilink
  • source
  • parent