LiveVal: Time-aware Data Valuation via Adaptive Reference Points

TL;DR

LiveVal is a novel, efficient, and adaptive data valuation method that integrates with training processes to detect harmful samples early, improving training efficiency and robustness across various models and data types.

Contribution

It introduces a real-time, reference-based valuation approach with adaptive reference points, addressing limitations of previous methods that rely on retraining or static assumptions.

Findings

Achieves 180x speedup over traditional data valuation methods.

Maintains robust detection of harmful samples across modalities.

Provides theoretical guarantees for stability and alignment with training progress.

Abstract

Time-aware data valuation enhances training efficiency and model robustness, as early detection of harmful samples could prevent months of wasted computation. However, existing methods rely on model retraining or convergence assumptions or fail to capture long-term training dynamics. We propose LiveVal, an efficient time-aware data valuation method with three key designs: 1) seamless integration with SGD training for efficient data contribution monitoring; 2) reference-based valuation with normalization for reliable benchmark establishment; and 3) adaptive reference point selection for real-time updating with optimized memory usage. We establish theoretical guarantees for LiveVal's stability and prove that its valuations are bounded and directionally aligned with optimization progress. Extensive experiments demonstrate that LiveVal provides efficient data valuation across different modalities and model scales, achieving 180 speedup over traditional methods while maintaining robust detection performance.