Search Results for author:
Found 24 papers, 6 papers with code
Learning to Watermark LLM-generated Text via Reinforcement Learning
While prior works focus on token-level watermark that embeds signals into the output, we design a model-level watermark that embeds signals into the LLM weights, and such signals can be detected by a paired detector.
Improving Reinforcement Learning from Human Feedback Using Contrastive Rewards
Reinforcement learning from human feedback (RLHF) is the mainstream paradigm used to align large language models (LLMs) with human preferences.
Fair Classifiers Without Fair Training: An Influence-Guided Data Sampling Approach
A fair classifier should ensure the benefit of people from different groups, while the group information is often sensitive and unsuitable for model training.
Measuring and Reducing LLM Hallucination without Gold-Standard Answers via Expertise-Weighting
In this work, we propose Factualness Evaluations via Weighting LLMs (FEWL), the first hallucination metric that is specifically designed for the scenario when gold-standard answers are absent.
Rethinking Machine Unlearning for Large Language Models
We explore machine unlearning (MU) in the domain of large language models (LLMs), referred to as LLM unlearning.
Human-Instruction-Free LLM Self-Alignment with Limited Samples
The key idea is to first retrieve high-quality samples related to the target domain and use them as In-context Learning examples to generate more samples.
Large Language Model Unlearning
To the best of our knowledge, our work is among the first to explore LLM unlearning.
Fair Classifiers that Abstain without Harm
To generalize the abstaining decisions to test samples, we then train a surrogate model to learn the abstaining decisions based on the IP solutions in an end-to-end manner.
Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment
However, a major challenge faced by practitioners is the lack of clear guidance on evaluating whether LLM outputs align with social norms, values, and regulations.
On the Cause of Unfairness: A Training Sample Perspective
Identifying the causes of a model's unfairness is an important yet relatively unexplored task.
Label Inference Attack against Split Learning under Regression Setting
In this work, we step further to study the leakage in the scenario of the regression model, where the private labels are continuous numbers (instead of discrete labels in classification).
Learning to Counterfactually Explain Recommendations
The key idea is to train a surrogate model to learn the effect of removing a subset of user history on the recommendation.
Weak Proxies are Sufficient and Preferable for Fairness with Missing Sensitive Attributes
Our theoretical analyses show that directly using proxy models can give a false sense of (un)fairness.
DPAUC: Differentially Private AUC Computation in Federated Learning
Federated learning (FL) has gained significant attention recently as a privacy-enhancing tool to jointly train a machine learning model by multiple participants.
Differentially Private Multi-Party Data Release for Linear Regression
Differentially Private (DP) data release is a promising technique to disseminate data without compromising the privacy of data subjects.
Differentially Private AUC Computation in Vertical Federated Learning
In this work, we propose two evaluation algorithms that can more accurately compute the widely used AUC (area under curve) metric when using label DP in vFL.
Differentially Private Label Protection in Split Learning
Split learning is a distributed training framework that allows multiple parties to jointly train a machine learning model over vertically partitioned data (partitioned by attributes).
Counterfactually Evaluating Explanations in Recommender Systems
Modern recommender systems face an increasing need to explain their recommendations.
Label Leakage and Protection from Forward Embedding in Vertical Federated Learning
As the raw labels often contain highly sensitive information, some recent work has been proposed to prevent the label leakage from the backpropagated gradients effectively in vFL.
Defending against Reconstruction Attack in Vertical Federated Learning
Recently researchers have studied input leakage problems in Federated Learning (FL) where a malicious party can reconstruct sensitive training inputs provided by users from shared gradient.
Vertical Federated Learning without Revealing Intersection Membership
In this paper, we propose a vFL framework based on Private Set Union (PSU) that allows each party to keep sensitive membership information to itself.
Backdoor Attacks Against Deep Learning Systems in the Physical World
A critical question remains unanswered: can backdoor attacks succeed using physical objects as triggers, thus making them a credible threat against deep learning systems in the real world?
Regula Sub-rosa: Latent Backdoor Attacks on Deep Neural Networks
Recent work has proposed the concept of backdoor attacks on deep neural networks (DNNs), where misbehaviors are hidden inside "normal" models, only to be triggered by very specific inputs.
Automated Crowdturfing Attacks and Defenses in Online Review Systems
Malicious crowdsourcing forums are gaining traction as sources of spreading misinformation online, but are limited by the costs of hiring and managing human workers.
Cryptography and Security Social and Information Networks
