Training on LLM text embeddings can cause tabular classifiers to collapse to single-class predictions, which spuriously inflates Accuracy-on-the-Line correlations.
Reassessing the Validity of Spurious Correlations Benchmarks
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Neural networks can fail when the data contains spurious correlations. To understand this phenomenon, researchers have proposed numerous spurious correlations benchmarks upon which to evaluate mitigation methods. However, we observe that these benchmarks exhibit substantial disagreement, with the best methods on one benchmark performing poorly on another. We explore this disagreement, and examine benchmark validity by defining three desiderata that a benchmark should satisfy in order to meaningfully evaluate methods. Our results have implications for both benchmarks and mitigations: we find that certain benchmarks are not meaningful measures of method performance, and that several methods are not sufficiently robust for widespread use. We present a simple recipe for practitioners to choose methods using the most similar benchmark to their given problem.
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Data Curation Matters: Model Collapse and Spurious Shift Performance Prediction from Training on Uncurated Text Embeddings
Training on LLM text embeddings can cause tabular classifiers to collapse to single-class predictions, which spuriously inflates Accuracy-on-the-Line correlations.