English
Related papers

Related papers: An Attribution Method for Siamese Encoders

200 papers

Sparse Autoencoders (SAEs) are a promising approach for extracting neural network representations by learning a sparse and overcomplete decomposition of the network's internal activations. However, SAEs are traditionally trained considering…

Machine Learning · Computer Science 2025-04-02 Jeffrey Olmo , Jared Wilson , Max Forsey , Bryce Hepner , Thomas Vin Howe , David Wingate

Measuring the semantic similarity between two sentences (or Semantic Textual Similarity - STS) is fundamental in many NLP applications. Despite the remarkable results in supervised settings with adequate labeling, little attention has been…

Computation and Language · Computer Science 2018-10-31 Xin Tang , Shanbo Cheng , Loc Do , Zhiyu Min , Feng Ji , Heng Yu , Ji Zhang , Haiqin Chen

Interpretable machine learning offers insights into what factors drive a certain prediction of a black-box system. A large number of interpreting methods focus on identifying explanatory input features, which generally fall into two main…

Machine Learning · Computer Science 2023-06-02 Vy Vo , Van Nguyen , Trung Le , Quan Hung Tran , Gholamreza Haffari , Seyit Camtepe , Dinh Phung

Using end-to-end models for speech translation (ST) has increasingly been the focus of the ST community. These models condense the previously cascaded systems by directly converting sound waves into translated text. However, cascaded models…

Computation and Language · Computer Science 2021-01-25 Orion Weller , Matthias Sperber , Christian Gollan , Joris Kluivers

Sparse autoencoders (SAEs) are a promising approach to extracting features from neural networks, enabling model interpretability as well as causal interventions on model internals. SAEs generate sparse feature representations using a…

Machine Learning · Computer Science 2024-11-11 Kola Ayonrinde

Image attribution analysis seeks to highlight the feature representations learned by visual models such that the highlighted feature maps can reflect the pixel-wise importance of inputs. Gradient integration is a building block in the…

Computer Vision and Pattern Recognition · Computer Science 2025-06-26 Róisín Luo , James McDermott , Colm O'Riordan

We propose a novel approach for spoofed speech characterization through explainable probabilistic attribute embeddings. In contrast to high-dimensional raw embeddings extracted from a spoofing countermeasure (CM) whose dimensions are not…

Audio and Speech Processing · Electrical Eng. & Systems 2024-09-18 Manasi Chhibber , Jagabandhu Mishra , Hyejin Shim , Tomi H. Kinnunen

Attribution maps are one of the most established tools to explain the functioning of computer vision models. They assign importance scores to input features, indicating how relevant each feature is for the prediction of a deep neural…

Computer Vision and Pattern Recognition · Computer Science 2024-12-10 Robin Hesse , Simone Schaub-Meyer , Stefan Roth

Feature attribution a.k.a. input salience methods which assign an importance score to a feature are abundant but may produce surprisingly different results for the same model on the same input. While differences are expected if disparate…

Computation and Language · Computer Science 2022-11-10 Jasmijn Bastings , Sebastian Ebert , Polina Zablotskaia , Anders Sandholm , Katja Filippova

Feature attribution methods aim to improve the transparency of deep neural networks by identifying the input features that influence a model's decision. Pixel-based heatmaps have become the standard for attributing features to…

Machine Learning · Statistics 2025-06-06 Gabriel Kasmi , Amandine Brunetto , Thomas Fel , Jayneel Parekh

Large language models (LLMs) make next-token predictions based on clues present in their context, such as semantic descriptions and in-context examples. Yet, elucidating which prior tokens most strongly influence a given prediction remains…

Computation and Language · Computer Science 2026-03-17 Toni J. B. Liu , Baran Zadeoğlu , Nicolas Boullé , Raphaël Sarfati , Christopher J. Earls

Pre-trained sentence embeddings have been shown to be very useful for a variety of NLP tasks. Due to the fact that training such embeddings requires a large amount of data, they are commonly trained on a variety of text data. An adaptation…

Computation and Language · Computer Science 2020-08-31 Anna Kruspe

Integrated Gradients (IG), a widely used axiomatic path-based attribution method, assigns importance scores to input features by integrating model gradients along a straight path from a baseline to the input. While effective in some cases,…

Machine Learning · Computer Science 2026-02-27 Sina Salek , Joseph Enguehard

Attribution scores indicate the importance of different input parts and can, thus, explain model behaviour. Currently, prompt-based models are gaining popularity, i.a., due to their easier adaptability in low-resource settings. However, the…

Computation and Language · Computer Science 2024-03-11 Wei Zhou , Heike Adel , Hendrik Schuff , Ngoc Thang Vu

Integrated gradients are widely employed to evaluate the contribution of input features in classification models because it satisfies the axioms for attribution of prediction. This method, however, requires an appropriate baseline for…

Machine Learning · Computer Science 2018-11-28 Kazuki Tachikawa , Yuji Kawai , Jihoon Park , Minoru Asada

Large Reasoning Models (LRMs) achieve strong reasoning performance by generating long chains of thought (CoTs), yet only a small fraction of these traces meaningfully contributes to answer prediction, while the majority contains repetitive…

Computation and Language · Computer Science 2026-02-03 Siyuan Wang , Yanchen Liu , Xiang Ren

Feature attribution methods, proposed recently, help users interpret the predictions of complex models. Our approach integrates feature attributions into the objective function to allow machine learning practitioners to incorporate priors…

Computation and Language · Computer Science 2019-06-21 Frederick Liu , Besim Avci

With the growing popularity of deep-learning models, model understanding becomes more important. Much effort has been devoted to demystify deep neural networks for better interpretability. Some feature attribution methods have shown…

Computation and Language · Computer Science 2022-04-27 Sheng Zhang , Jin Wang , Haitao Jiang , Rui Song

Traditional methods for spatial inference estimate smooth interpolating fields based on features measured at well-located points. When the spatial locations of some observations are missing, joint inference of the fields and locations is…

Semantic sentence embedding models encode natural language sentences into vectors, such that closeness in embedding space indicates closeness in the semantics between the sentences. Bilingual data offers a useful signal for learning such…

Computation and Language · Computer Science 2020-11-20 John Wieting , Graham Neubig , Taylor Berg-Kirkpatrick