paper-with-me

홈 › Papers

Signature Activation: A Sparse Signal View for Holistic Saliency

2023-09-20 · Jose Roberto Tello Ayala, Akl C. Fahed, Weiwei Pan, Eugene V. Pomerantsev, Patrick T. Ellinor, Anthony Philippakis, Finale Doshi-Velez

The adoption of machine learning in healthcare calls for model transparency and explainability. In this work, we introduce Signature Activation, a saliency method that generates holistic and class-agnostic explanations for Convolutional Neural Network (CNN) outputs. Our method exploits the fact that certain kinds of medical images, such as angiograms, have clear foreground and background objects. We give theoretical explanation to justify our methods. We show the potential use of our method in clinical settings through evaluating its efficacy for aiding the detection of lesions in coronary angiograms.

📄 PDF Abstract BibTeX arXiv:2309.11443

Code (1)

dtak/signature-activation 공식 구현

Similar Papers 제목 키워드 기반

Task-Conditioned Routing Signatures in Sparse Mixture-of-Experts Transformers

2026-03-11 · Mynampati Sri Ranganadha Avinash arxiv

Sparse Mixture-of-Experts (MoE) architectures enable efficient scaling of large language models through conditional computation, yet the routing mechanisms responsible for expert selection remain poorly understood. In th…

MiA-Signature: Approximating Global Activation for Long-Context Understanding

2026-05-07 · Yuqing Li, Jiangnan Li, Mo Yu, Zheng Lin 외 arxiv

A growing body of work in cognitive science suggests that reportable conscious access is associated with \emph{global ignition} over distributed memory systems, while such activation is only partially accessible as indiv…

Long-Context Understanding

LLM Self-Recognition: Steering and Retrieving Activation Signatures

2026-06-04 · Thibaud Ardoin, Jonas Schäfer, Gerhard Wunder arxiv

Recent advances in interpretability suggest that large language models (LLMs) implicitly encode signals in their generated text that enable self-recognition of their outputs. We demonstrate that this capability is reliab…

Sparse maximal update parameterization: A holistic approach to sparse training dynamics

2024-05-24 · Nolan Dey, Shane Bergsma, Joel Hestness

Several challenges make it difficult for sparse neural networks to compete with dense models. First, setting a large fraction of weights to zero impairs forward and gradient signal propagation. Second, sparse studies oft…

Language ModelingLanguage Modelling

Pre-Intervention Prediction of Sparse Autoencoder Steering Side Effects

2026-06-06 · Evan Duan arxiv

Sparse autoencoder (SAE) features are increasingly used to steer language models, but feature steering is rarely clean: the same intervention can behave inconsistently across contexts and perturb unrelated features. We i…