6 papers
Generating Risky Samples with Conformity Constraints via Diffusion Models
Han Yu, Hao Zou, Xingxuan Zhang +4
Although neural networks achieve promising performance in many tasks, they may still fail when encountering some examples and bring about risks to applications. To discover risky s…
ODP-Bench: Benchmarking Out-of-Distribution Performance Prediction
Han Yu, Kehan Li, Dongbai Li +3
Recently, there has been gradually more attention paid to Out-of-Distribution (OOD) performance prediction, whose goal is to predict the performance of trained models on unlabeled…
COUNTS: Benchmarking Object Detectors and Multimodal Large Language Models under Distribution Shifts
Jiansheng Li, Xingxuan Zhang, Hao Zou +6
Current object detectors often suffer significant perfor-mance degradation in real-world applications when encountering distributional shifts. Consequently, the out-of-distribution…
Understanding the Generalization of In-Context Learning in Transformers: An Empirical Study
Xingxuan Zhang, Haoran Wang, Jiansheng Li +6
Large language models (LLMs) like GPT-4 and LLaMA-3 utilize the powerful in-context learning (ICL) capability of Transformer architecture to learn on the fly from limited examples.…
Sample Weight Averaging for Stable Prediction
Han Yu, Yue He, Renzhe Xu +4
The challenge of Out-of-Distribution (OOD) generalization poses a foundational concern for the application of machine learning algorithms to risk-sensitive areas. Inspired by tradi…
Error Slice Discovery via Manifold Compactness
Han Yu, Hao Zou, Jiashuo Liu +4
Despite the great performance of deep learning models in many areas, they still make mistakes and underperform on certain subsets of data, i.e. error slices. Given a trained model,…