Picture for Huiqi Deng

Huiqi Deng

Towards Attributions of Input Variables in a Coalition

Add code
Sep 23, 2023
Viaarxiv icon

Mitigating Shortcuts in Language Models with Soft Label Encoding

Add code
Sep 17, 2023
Viaarxiv icon

Explainability for Large Language Models: A Survey

Add code
Sep 17, 2023
Viaarxiv icon

Understanding and Unifying Fourteen Attribution Methods with Taylor Interactions

Add code
Mar 06, 2023
Viaarxiv icon

Bayesian Neural Networks Tend to Ignore Complex and Sensitive Concepts

Add code
Feb 25, 2023
Viaarxiv icon

Concept-Level Explanation for the Generalization of a DNN

Add code
Feb 25, 2023
Viaarxiv icon

Towards Axiomatic, Hierarchical, and Symbolic Explanation for Deep Models

Add code
Nov 30, 2021
Figure 1 for Towards Axiomatic, Hierarchical, and Symbolic Explanation for Deep Models
Figure 2 for Towards Axiomatic, Hierarchical, and Symbolic Explanation for Deep Models
Figure 3 for Towards Axiomatic, Hierarchical, and Symbolic Explanation for Deep Models
Figure 4 for Towards Axiomatic, Hierarchical, and Symbolic Explanation for Deep Models
Viaarxiv icon

Discovering and Explaining the Representation Bottleneck of DNNs

Add code
Nov 18, 2021
Figure 1 for Discovering and Explaining the Representation Bottleneck of DNNs
Figure 2 for Discovering and Explaining the Representation Bottleneck of DNNs
Figure 3 for Discovering and Explaining the Representation Bottleneck of DNNs
Figure 4 for Discovering and Explaining the Representation Bottleneck of DNNs
Viaarxiv icon

A General Taylor Framework for Unifying and Revisiting Attribution Methods

Add code
May 28, 2021
Figure 1 for A General Taylor Framework for Unifying and Revisiting Attribution Methods
Figure 2 for A General Taylor Framework for Unifying and Revisiting Attribution Methods
Figure 3 for A General Taylor Framework for Unifying and Revisiting Attribution Methods
Figure 4 for A General Taylor Framework for Unifying and Revisiting Attribution Methods
Viaarxiv icon

Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution

Add code
Apr 14, 2021
Figure 1 for Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution
Figure 2 for Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution
Figure 3 for Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution
Figure 4 for Mutual Information Preserving Back-propagation: Learn to Invert for Faithful Attribution
Viaarxiv icon