paper-with-me

Papers

Grad-SAM: Explaining Transformers via Gradient Self-Attention Maps

2022-04-23 · Oren Barkan, Edan Hauon, Avi Caciularu, Ori Katz, Itzik Malkiel, Omri Armstrong, Noam Koenigstein

Transformer-based language models significantly advanced the state-of-the-art in many linguistic tasks. As this revolution continues, the ability to explain model predictions has become a major area of interest for the NLP community. In this work, we present Gradient Self-Attention Maps (Grad-SAM) - a novel gradient-based method that analyzes self-attention units and identifies the input elements that explain the model's prediction the best. Extensive evaluations on various benchmarks show that Grad-SAM obtains significant improvements over state-of-the-art alternatives.

📄 PDF Abstract BibTeX arXiv:2204.11073

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Gradient Flow Polarizes Softmax Outputs towards Low-Entropy Solutions

2026-03-06 · Aditya Varre, Mark Rofin, Nicolas Flammarion arxiv

Understanding the intricate non-convex training dynamics of softmax-based models is crucial for explaining the empirical success of transformers. In this article, we analyze the gradient flow dynamics of the value-softma…

APRIL: Finding the Achilles' Heel on Privacy for Vision Transformers

2021-12-28 · CVPR 2022 1 · Jiahao Lu, Xi Sheryl Zhang, Tianli Zhao, Xiangyu He 외

Federated learning frameworks typically require collaborators to share their local gradient updates of a common model instead of sharing training data to preserve privacy. However, prior works on Gradient Leakage Attacks…

Federated Learning

How Transformers Learn Causal Structure with Gradient Descent

2024-02-22 · Eshaan Nichani, Alex Damian, Jason D. Lee

The incredible success of transformers on sequence modeling tasks can be largely attributed to the self-attention mechanism, which allows information to be transferred between different parts of a sequence. Self-attentio…

In-Context Learning

Transformers learn in-context by gradient descent

2022-12-15 · Johannes von Oswald, Eyvind Niklasson, Ettore Randazzo, João Sacramento 외

At present, the mechanisms of in-context learning in Transformers are not well understood and remain mostly an intuition. In this paper, we suggest that training Transformers on auto-regressive objectives is closely rela…

In-Context LearningMeta-Learningregression

Transformers Efficiently Perform In-Context Logistic Regression via Normalized Gradient Descent

2026-05-07 · Chenyang Zhang, Yuan Cao arxiv

Transformers have demonstrated remarkable in-context learning (ICL) capabilities. The strong ICL performance of transformers is commonly believed to arise from their ability to implicitly execute certain algorithms on th…