Cover Image

How Faithful Is Trajectory-Based Data Attribution? Error Sources, Remedies, and Practical Guidelines

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Deng, Junwei, Hu, Pingbang, Jin, Suliang, Lu, Hao, Wang, Jiachen T., Zhang, Shichang, Ma, Jiaqi W.
Format:	Preprint
Published:	2026
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Adversarial Attacks on Data Attribution
by: Wang, Xinhe, et al.
Published: (2024)

Efficient Ensembles Improve Training Data Attribution
by: Deng, Junwei, et al.
Published: (2024)

$\texttt{dattri}$: A Library for Efficient Data Attribution
by: Deng, Junwei, et al.
Published: (2024)

GraSS: Scalable Data Attribution with Gradient Sparsification and Sparse Projection
by: Hu, Pingbang, et al.
Published: (2025)

A Versatile Influence Function for Data Attribution with Non-Decomposable Loss
by: Deng, Junwei, et al.
Published: (2024)

A Reliable Cryptographic Framework for Empirical Machine Unlearning Evaluation
by: Tu, Yiwen, et al.
Published: (2024)

Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining
by: Wang, Weiyi, et al.
Published: (2025)

Most Influential Subset Selection: Challenges, Promises, and Beyond
by: Hu, Yuzheng, et al.
Published: (2024)

Exploring Training Data Attribution under Limited Access Constraints
by: Zhang, Shiyuan, et al.
Published: (2025)

Dr. Post-Training: A Data Regularization Perspective on LLM Post-Training
by: Hu, Pingbang, et al.
Published: (2026)

A Unified Theory of Random Projection for Influence Functions
by: Hu, Pingbang, et al.
Published: (2026)

Pseudo-Nonlinear Data Augmentation: A Constrained Energy Minimization Viewpoint
by: Hu, Pingbang, et al.
Published: (2024)

Who Gets Credit or Blame? Attributing Accountability in Modern AI Systems
by: Zhang, Shichang, et al.
Published: (2025)

Generalized Group Data Attribution
by: Ley, Dan, et al.
Published: (2024)

OATS: Online Data Augmentation for Time Series Foundation Models
by: Deng, Junwei, et al.
Published: (2026)

Towards Unified Attribution in Explainable AI, Data-Centric AI, and Mechanistic Interpretability
by: Zhang, Shichang, et al.
Published: (2025)

ABE: A Unified Framework for Robust and Faithful Attribution-Based Explainability
by: Zhu, Zhiyu, et al.
Published: (2025)

Detecting and Filtering Unsafe Training Data via Data Attribution with Denoised Representation
by: Pan, Yijun, et al.
Published: (2025)

Daunce: Data Attribution through Uncertainty Estimation
by: Pan, Xingyuan, et al.
Published: (2025)

Measuring Fine-Grained Relatedness in Multitask Learning via Data Attribution
by: Tu, Yiwen, et al.
Published: (2025)

Increasing Stability for the Inverse Source Problem of the Schrödinger Operator
by: Suliang Si
Published: (2026)

AttributionLab: Faithfulness of Feature Attribution Under Controllable Environments
by: Zhang, Yang, et al.
Published: (2023)

Wrapper Boxes: Faithful Attribution of Model Predictions to Training Data
by: Su, Yiheng, et al.
Published: (2023)

A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning
by: Hu, Yuzheng, et al.
Published: (2025)

Practical Attribution Guidance for Rashomon Sets
by: Li, Sichao, et al.
Published: (2024)

Ideal Attribution and Faithful Watermarks for Language Models
by: Song, Min Jae, et al.
Published: (2025)

Source Attribution for Large Language Model-Generated Data
by: Wang, Jingtan, et al.
Published: (2023)

Natural Geometry of Robust Data Attribution: From Convex Models to Deep Networks
by: Li, Shihao, et al.
Published: (2025)

Incorporating Attribution Importance for Improving Faithfulness Metrics
by: Zhao, Zhixue, et al.
Published: (2023)

DOCTOR: Dynamic On-Chip Temporal Variation Remediation Toward Self-Corrected Photonic Tensor Accelerators
by: Lu, Haotian, et al.
Published: (2024)

Faithful or Just Plausible? Evaluating the Faithfulness of Closed-Source LLMs in Medical Reasoning
by: Afolabi, Halimat, et al.
Published: (2026)

Semantic Trajectory Data Mining with LLM-Informed POI Classification
by: Liu, Yifan, et al.
Published: (2024)

Beyond Output Faithfulness: Learning Attributions that Preserve Computational Pathways
by: Zhang, Siyu, et al.
Published: (2025)

MemTrace: Tracing and Attributing Errors in Large Language Model Memory Systems
by: Deng, Xinle, et al.
Published: (2026)

Conformal Agent Error Attribution
by: Feng, Naihe, et al.
Published: (2026)

How to Upscale Neural Networks with Scaling Law? A Survey and Practical Guidelines
by: Sengupta, Ayan, et al.
Published: (2025)

Normalized AOPC: Fixing Misleading Faithfulness Metrics for Feature Attribution Explainability
by: Edin, Joakim, et al.
Published: (2024)

Towards Long-Horizon Interpretability: Efficient and Faithful Multi-Token Attribution for Reasoning LLMs
by: Pan, Wenbo, et al.
Published: (2026)

Faithfulness Evaluation for Decoder-only LLM Attributions with Controlled Retained Information
by: Huang, Xin, et al.
Published: (2026)

From Indirect Object Identification to Syllogisms: Exploring Binary Mechanisms in Transformer Circuits
by: Saraipour, Karim, et al.
Published: (2025)