← Search

Shi-Yu Tian

2 accepted papers

2026

On the Learnability of Test-Time Adaptation: A Recovery Complexity Perspective

ICML 2026poster

Test-time adaptation (TTA) aims to adapt models to maintain reliable performance on non-stationary test streams without requiring labeled data. Despite its empirical success, the learnability of TTA under distributional non-stationarity remains unexplored. A key challenge is lacking of a principled …

Cited by 0SourceScholar
2025

VCSearch: Bridging the Gap Between Well-Defined and Ill-Defined Problems in Mathematical Reasoning

EMNLP 2025

Large language models (LLMs) have demonstrated impressive performance on reasoning tasks, including mathematical reasoning. However, the current evaluation mostly focuses on carefully constructed benchmarks and neglects the consideration of real-world reasoning problems that present missing or contr

Cited by 0SourcePDFScholar