computer-vision-4

Network Interpretation

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in network-interpretation-4

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find network-interpretation-4 models and implementations

Datasets

No datasets available.

Subtasks

No subtasks available.

Most implemented papers

Plug-and-Play VQA: Zero-shot VQA by Conjoining Large Pretrained Models with Zero Training

S. Hoi, S. Savarese, Junnan Li, Boyang Albert Li, A. Tiong•Sun Oct 16 2022

This work first generate question-guided informative image captions, and pass the captions to a PLM as context for question answering, and achieves state-of-the-art results on zero-shot VQAv2 and GQA.

132

Content

0

Paper Graph

Fooling Neural Network Interpretations via Adversarial Model Manipulation

Taesup Moon, Juyeon Heo, Sunghwan Joo•Thu Jan 31 2019

It is claimed that the stability of neural network interpretation method with respect to the authors' adversarial model manipulation is an important criterion to check for developing robust and reliable neural network interpretations method.

225 0

Paper Graph

An Empirical Study on the Relation Between Network Interpretability and Adversarial Robustness

D. Dou, Boyang Albert Li, Adam Noack, Isaac Ahern•Sat Dec 05 2020

It is demonstrated that training the networks to have interpretable gradients improves their robustness to adversarial perturbations, especially in cross-norm attacks and under heavy perturbation.

34 0

Paper Graph

Proper Network Interpretability Helps Adversarial Robustness in Classification

Pin-Yu Chen, Shiyu Chang, Sijia Liu, Akhilan Boopathy, Gaoyuan Zhang, Cynthia Liu, L. Daniel•Thu Jun 25 2020

This paper theoretically shows that with a proper measurement of interpretation, it is actually difficult to prevent prediction-evasion adversarial attacks from causing interpretation discrepancy, and develops an interpretability-aware defensive scheme built only on promoting robust interpretation (without the need for resorting to adversarial loss minimization).

73 0

Paper Graph

Attribution Preservation in Network Compression for Reliable Network Interpretation

Eunho Yang, Sung Ju Hwang, Geondo Park, J. Yang•Tue Oct 27 2020

This paper presents a framework that can preserve the attributions while compressing a network by employing the Weighted Collapsed Attribution Matching regularizer, and demonstrates the effectiveness of the algorithm both quantitatively and qualitatively on diverse compression methods.

7 0

Paper Graph

Towards More Robust Interpretation via Local Gradient Alignment

Taesup Moon, Juyeon Heo, Sunghwan Joo, Seokhyeon Jeong, Adrian Weller•Mon Nov 28 2022

This work forms a normalization invariant cosine distance-based criterion and derives its upper bound, which gives insight for why simply minimizing the Hessian norm at the input, as has been done in previous work, is not sufficient for attaining robust feature attribution.

7 0

Paper Graph

Adding a benchmark result helps the community track progress.