Distributionally Robust Set Representation Learning Under Inference-Time Element Corruption

Standard Set Representation Learning methods typically excel on curated data but often overlook the challenge of Inference-time Element Corruption. This refers to scenarios where deployed models encounter element-level degradations, such as outliers or missing components, that may distort set representation and degrade performance. We propose SW-DRSO, a distributionally robust optimization framework tailored for sets. Rather than minimizing loss solely on observed training data, SW-DRSO optimizes a tractable surrogate of the worst-case expected loss over a family of plausible inference-time variations. We introduce a barycentric adversary that approximates the intractable search over corrupted sets by a differentiable training-time optimization over simplex weights. Extensive experiments across four tasks demonstrate that SW-DRSO effectively enhances robustness against corruption while maintaining high overall performance.

We propose SW-DRSO, a distributionally robust optimization framework that tackles inference-time element corruption in set representation learning. By introducing a barycentric adversary, it optimizes a tractable surrogate of the worst-case expected loss through differentiable training-time optimization. Experiments prove that SW-DRSO significantly enhances robustness against element-level degradations while maintaining high performance.