Search Results for author:
Found 11 papers, 8 papers with code
Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models
Our method and dataset can help developers evaluate and fine-tune safer and more usable LLMs.
Can Watermarking Large Language Models Prevent Copyrighted Text Generation and Hide Training Data?
Surprisingly, we find that watermarking adversely affects the success rate of MIAs, complicating the task of detecting copyrighted text in the pretraining dataset.
Calibrated Dataset Condensation for Faster Hyperparameter Search
We aim to generate a synthetic validation dataset so that the validation-performance rankings of the models, with different hyperparameters, on the condensed and original datasets are comparable.
Shadowcast: Stealthy Data Poisoning Attacks Against Vision-Language Models
We show that Shadowcast are highly effective in achieving attacker's intentions using as few as 50 poison samples.
Mementos: A Comprehensive Benchmark for Multimodal Large Language Model Reasoning over Image Sequences
However, current MLLM benchmarks are predominantly designed to evaluate reasoning based on static information about a single image, and the ability of modern MLLMs to extrapolate from image sequences, which is essential for understanding our ever-changing world, has been less investigated.
WAVES: Benchmarking the Robustness of Image Watermarks
Our evaluation examines two pivotal dimensions: the degree of image quality degradation and the efficacy of watermark detection after attacks.
C-Disentanglement: Discovering Causally-Independent Generative Factors under an Inductive Bias of Confounder
Representation learning assumes that real-world data is generated by a few semantically meaningful generative factors (i. e., sources of variation) and aims to discover them in the latent space.
Adapting Static Fairness to Sequential Decision-Making: Bias Mitigation Strategies towards Equal Long-term Benefit Rate
To address biases in sequential decision-making, we introduce a long-term fairness concept named Equal Long-term Benefit Rate (ELBERT).
Exploring and Exploiting Decision Boundary Dynamics for Adversarial Robustness
However, it is unclear whether existing robust training methods effectively increase the margin for each vulnerable point during training.
Multi-Task Adversarial Attack
More specifically, MTA uses a generator for adversarial perturbations which consists of a shared encoder for all tasks and multiple task-specific decoders.
FREEtree: A Tree-based Approach for High Dimensional Longitudinal Data With Correlated Features
This paper proposes FREEtree, a tree-based method for high dimensional longitudinal data with correlated features.
