Transformer-Explainability
Transformer-Explainability copied to clipboard
hila-chefer
[CVPR 2021] Official PyTorch implementation for Transformer Interpretability Beyond Attention Visualization, a novel method to visualize classifications by Transformer based networks.
Hello , First, thank you for this great work. It definitely pushes the xai for transformer to a new frontier, and it inspires many people (including me) to work in...
While attempting to utilize this repo for my research purposes, I realized that there were some missing imports that resulted in frequent not found errors
Hi Hila, Outstanding Work!! I'm thinking to implement LRP for `mean` & `sum` function. How would you suggest implementing it? I'm thinking to use `Linear` layer if I know the...
HI Chefer, Is there a typo with Lines 218 to 219 https://github.com/hila-chefer/Transformer-Explainability/blob/c3e578f76b954e8528afeaaee26de3f07e3fe559/modules/layers_ours.py#L218-L219 which should be the below? `S1 = safe_divide(R, Z1)` `S2 = safe_divide(R, Z2)` according to https://github.com/wjNam/Relative_Attributing_Propagation/blob/7fa96822740591b605712f251e556ec8487d1eea/modules/layers.py#L268-L269C36 Thanks, Hongbo
The generic Deep Taylor Decomposition formula in the paper seems to be different as the formula in reference 27. The Deep Taylor Decomposition formula in Reference 27 requires selecting a...
when I run the code I get an error on the explanation generator. How can I solve it?
Hi, first of all thanks for your research and your well prepared code! While implementing the code for my own custom model, I noticed that for linear layers, the relprop...
Existing code used the lrp instance instead of orig_lrp, which led to the same attribution as the transformer attribution. Following this change the produced attributions are different as expected. Also...
# Patching CVE-2007-4559 Hi, we are security researchers from the Advanced Research Center at [Trellix](https://www.trellix.com). We have began a campaign to patch a widespread bug named CVE-2007-4559. CVE-2007-4559 is a...
Metadata
Owner
Metadata
[CVPR 2021] Official PyTorch implementation for Transformer Interpretability Beyond Attention Visualization, a novel method to visualize classifications by Transformer based networks.