Imagen de Portada

Systematic Evaluation of Attribution Methods: Eliminating Threshold Bias and Revealing Method-Dependent Performance Patterns

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autor principal:	Aksoy, Serra
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Benchmarking Vision Transformers and CNNs for Thermal Photovoltaic Fault Detection with Explainable AI Validation
por: Aksoy, Serra
Publicado: (2025)

Better Understanding Differences in Attribution Methods via Systematic Evaluations
por: Rao, Sukrut, et al.
Publicado: (2023)

Exploring the Relationship Between Feature Attribution Methods and Model Performance
por: Silva, Priscylla, et al.
Publicado: (2024)

Evaluating Feature Attribution Methods in the Image Domain
por: Gevaert, Arne, et al.
Publicado: (2022)

A Dual-Perspective Approach to Evaluating Feature Attribution Methods
por: Li, Yawei, et al.
Publicado: (2023)

Should Bias be Eliminated? A General Framework to Use Bias for OOD Generalization
por: Li, Yan, et al.
Publicado: (2025)

BOSE: A Systematic Evaluation Method Optimized for Base Models
por: Luan, Hongzhi, et al.
Publicado: (2025)

Parallel Algorithm for Optimal Threshold Labeling of Ordinal Regression Methods
por: Yamasaki, Ryoya, et al.
Publicado: (2024)

Remarks on Loss Function of Threshold Method for Ordinal Regression Problem
por: Yamasaki, Ryoya, et al.
Publicado: (2024)

On the Consistency of Kernel Methods with Dependent Observations
por: Massiani, Pierre-François, et al.
Publicado: (2024)

A Robust and Non-Iterative Tensor Decomposition Method with Automatic Thresholding
por: Hasegawa, Hiroki, et al.
Publicado: (2025)

Interaction Concordance Index: Performance Evaluation for Interaction Prediction Methods
por: Pahikkala, Tapio, et al.
Publicado: (2025)

Gender Bias in Explainability: Investigating Performance Disparity in Post-hoc Methods
por: Dhaini, Mahdi, et al.
Publicado: (2025)

Backward Compatibility in Attributive Explanation and Enhanced Model Training Method
por: Matsuno, Ryuta
Publicado: (2024)

Comparing Methods for Bias Mitigation in Graph Neural Networks
por: Hoffmann, Barbara, et al.
Publicado: (2025)

Pairwise Elimination with Instance-Dependent Guarantees for Bandits with Cost Subsidy
por: Juneja, Ishank, et al.
Publicado: (2025)

Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods?
por: Chen, Zihan, et al.
Publicado: (2025)

Class-Dependent Perturbation Effects in Evaluating Time Series Attributions
por: Baer, Gregor, et al.
Publicado: (2025)

Missingness Bias Calibration in Feature Attribution Explanations
por: Sridhar, Shailesh, et al.
Publicado: (2026)

Can Global XAI Methods Reveal Injected Bias in LLMs? SHAP vs Rule Extraction vs RuleSHAP
por: Sovrano, Francesco
Publicado: (2025)

Towards Better Understanding Attribution Methods
por: Rao, Sukrut, et al.
Publicado: (2022)

Adaptive Threshold-Driven Continuous Greedy Method for Scalable Submodular Optimization
por: Rostami, Mohammadreza, et al.
Publicado: (2026)

xaitimesynth: A Python Package for Evaluating Attribution Methods for Time Series with Synthetic Ground Truth
por: Baer, Gregor
Publicado: (2026)

Eliminating Inductive Bias in Reward Models with Information-Theoretic Guidance
por: Li, Zhuo, et al.
Publicado: (2025)

Eliminating Position Bias of Language Models: A Mechanistic Approach
por: Wang, Ziqi, et al.
Publicado: (2024)

Disentangling Interactions and Dependencies in Feature Attribution
por: König, Gunnar, et al.
Publicado: (2024)

FreqX: Analyze the Attribution Methods in Another Domain
por: Liu, Zechen, et al.
Publicado: (2024)

Attribution Methods in Asset Pricing: Do They Account for Risk?
por: Chen, Dangxing, et al.
Publicado: (2024)

Rethinking Robustness: A New Approach to Evaluating Feature Attribution Methods
por: Kiourti, Panagiota, et al.
Publicado: (2025)

Eliminating Domain Bias for Federated Learning in Representation Space
por: Zhang, Jianqing, et al.
Publicado: (2023)

ChaosMining: A Benchmark to Evaluate Post-Hoc Local Attribution Methods in Low SNR Environments
por: Shi, Ge, et al.
Publicado: (2024)

An Attribute-based Method for Video Anomaly Detection
por: Reiss, Tal, et al.
Publicado: (2022)

When is the Computation of a Feature Attribution Method Tractable?
por: Barceló, P., et al.
Publicado: (2025)

Integrated Gradient Correlation: a Dataset-wise Attribution Method
por: Lelièvre, Pierre, et al.
Publicado: (2024)

Conditional KRR: Injecting Unpenalized Features into Kernel Methods with Applications to Kernel Thresholding
por: Takhanov, Rustem, et al.
Publicado: (2026)

Final-Model-Only Data Attribution with a Unifying View of Gradient-Based Methods
por: Wei, Dennis, et al.
Publicado: (2024)

Moderate Actor-Critic Methods: Controlling Overestimation Bias via Expectile Loss
por: Hwang, Ukjo, et al.
Publicado: (2025)

Path Choice Matters for Clear Attribution in Path Methods
por: Zhang, Borui, et al.
Publicado: (2024)

GraphFramEx: Towards Systematic Evaluation of Explainability Methods for Graph Neural Networks
por: Amara, Kenza, et al.
Publicado: (2022)

Extending Fair Null-Space Projections for Continuous Attributes to Kernel Methods
por: Störck, Felix, et al.
Publicado: (2025)