Law
A Systematic Review of NeurIPS Dataset Management Practices
Datasets serve as a fundamental bedrock for machine learning models.
Bag of Tricks: Benchmarking of Jailbreak Attacks on LLMs
To address these issues, we introduced JailTrickBench to evaluate the impact of various attack settings on LLM performance and provide a baseline for jailbreak attacks, encouraging the adoption of a standardized evaluation framework.
Supplementary Materials: FiV A: Fine-grained Visual Attribute Dataset for T ext-to-Image Diffusion Models
Section A. We then introduce additional details on dataset construction in Section B. Further, we Finally, we discuss the limitations and future work of the project in Section D. Please also find the Details on attribute taxonomy and statistics. We visualize the rough distribution of visual attributes and subjects on the left. We also visualize the attribute alignment accuracy via human validation here. Due to space limitations, only 15 sub-subjects are listed for each major-subject. The result shows that Image 4 exhibits inconsistencies, with the reasons provided.
FiV A: Fine-grained Visual Attribute Dataset for Text-to-Image Diffusion Models Tong Wu
Recent advances in text-to-image generation have enabled the creation of high-quality images with diverse applications.
Toward a Well-Calibrated Discrimination via Survival Outcome-A ware Contrastive Learning
Previous deep learning approaches for survival analysis have primarily relied on ranking losses to improve discrimination performance, which often comes at the expense of calibration performance. To address such an issue, we propose a novel contrastive learning approach specifically designed to enhance discrimination without sacrificing calibration.
