Layer-Aware Influence for Online Data Valuation Estimation

ICLR 2026 Conference Submission13379 Authors

Keywords: Data-centric learning, Online Data Valuation Estimation

Abstract: Data-centric learning emphasizes curating high-quality training samples to boost performance rather than designing new architectures. A central problem is to estimate the influence of training sample efficiently. Prior studies largely focus on static influence measured on a converged model, overlooking how data valuation dynamically changes during optimization. This omission neglects the dynamic nature of sample influence during optimization, especially in deep models. To address the computational burden of frequent influence estimation, we develop a layer-aware online estimator that requires only loss-to-output gradients. This design avoids parameter-level and full-network gradients while preserving ranking fidelity. Extensive experiments across LLM pretraining, fine-tuning and image classification demonstrate that our method improves accuracy with substantially lower time and memory cost in both text and image datasets, making dynamic data curation both efficient and scalable in practice.

Primary Area: other topics in machine learning (i.e., none of the above)

Submission Number: 13379