passdrill
AI & LLM Engineering · Fine-tuning & Model Customisation · Card 006/010 hard

Both bottleneck adapter modules (Houlsby et al.) and LoRA are parameter-efficient fine-tuning methods that add a small number of new trainable parameters to a frozen pretrained model, but they differ in how those new parameters interact with the model's forward pass at inference time. Which statement correctly distinguishes them on this specific point?

  1. Bottleneck adapters add their new parameters as a parallel low-rank update to existing weight matrices that can be merged back into those weights after training, while LoRA inserts new down-projection and up-projection feed-forward layers with a nonlinearity directly into the forward pass
  2. Both methods add their new parameters in mathematically identical ways, differing only in what random distribution is used to initialize the new weights before training begins
  3. Both methods modify the exact same weight matrices inside the attention mechanism in an identical fashion, differing only in the name each paper's authors chose to give the technique
  4. Bottleneck adapters insert new down-projection and up-projection feed-forward layers with a nonlinearity directly into the forward pass, adding extra sequential computation and inference latency, while LoRA's trainable matrices form a parallel low-rank update to an existing weight matrix that can be merged back into that weight after training, adding no extra inference latency
Next card → Shuffle