Search Results for author:
Found 25 papers, 14 papers with code
Shape it Up! Restoring LLM Safety during Finetuning
Finetuning large language models (LLMs) enables user-specific customization but introduces critical safety risks: even a few harmful examples can compromise safety alignment.
Llama Guard 3-1B-INT4: Compact and Efficient Safeguard for Human-AI Conversations
This paper presents Llama Guard 3-1B-INT4, a compact and efficient Llama Guard model, which has been open-sourced to the community during Meta Connect 2024.
Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations
We introduce Llama Guard 3 Vision, a multimodal LLM-based safeguard for human-AI conversations that involves image understanding: it can be used to safeguard content for both multimodal LLM inputs (prompt classification) and outputs (response classification).
Towards Understanding the Fragility of Multilingual LLMs against Fine-Tuning Attacks
We first discover cross-lingual generalization of fine-tuning attacks: using a few adversarially chosen instruction-following examples in one language, multilingual LLMs can also be easily compromised (e. g., multilingual LLMs fail to refuse harmful prompts in other languages).
Persistent Pre-Training Poisoning of LLMs
Our work evaluates for the first time whether language models can also be compromised during pre-training, with a focus on the persistence of pre-training attacks after models are fine-tuned as helpful and harmless chatbots (i. e., after SFT and DPO).
Backtracking Improves Generation Safety
In the context of language model safety, when a partial unsafe generation is produced, language models by their nature tend to happily keep on generating similarly unsafe additional text.
BadMerging: Backdoor Attacks Against Model Merging
Considering that a merged model may incorporate tasks from different domains, BadMerging can jointly compromise the tasks provided by the adversary (on-task attack) and other contributors (off-task attack) and solve the corresponding unique challenges with novel attack designs.
The Llama 3 Herd of Models
This paper presents a new set of foundation models, called Llama 3.
Ranked #3 on
Question Answering
on PeerQA
EAVE: Efficient Product Attribute Value Extraction via Lightweight Sparse-layer Interaction
Specifically, we employ a heavy encoder to separately encode the product context and attribute.
Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
We introduce Llama Guard, an LLM-based input-output safeguard model geared towards Human-AI conversation use cases.
Where have you been? A Study of Privacy Risk for Point-of-Interest Recommendation
To better understand and quantify the privacy leakage in mobility data-based ML models, we design a privacy attack suite containing data extraction and membership inference attacks tailored for point-of-interest (POI) recommendation models, one of the most widely used mobility data-based ML models.
FFB: A Fair Fairness Benchmark for In-Processing Group Fairness Methods
This paper introduces the Fair Fairness Benchmark (\textsf{FFB}), a benchmarking framework for in-processing group fairness methods.
PLUE: Language Understanding Evaluation Benchmark for Privacy Policies in English
Privacy policies provide individuals with information about their rights and how their personal information is handled.
Reliable Representations Make A Stronger Defender: Unsupervised Structure Refinement for Robust GNN
However, both strategies are faced with some immediate problems: raw features cannot represent various properties of nodes (e. g., structure information), and representations learned by supervised GNN may suffer from the poor performance of the classifier on the poisoned graph.
Conditional Supervised Contrastive Learning for Fair Text Classification
Contrastive representation learning has gained much attention due to its superior performance in learning representations from both image and sequential data.
Retrieval Enhanced Data Augmentation for Question Answering on Privacy Policies
Prior studies in privacy policies frame the question answering (QA) task as identifying the most relevant text segment or a list of sentences from a policy document given a user query.
Towards Return Parity in Markov Decision Processes
We first provide a decomposition theorem for return disparity, which decomposes the return disparity of any two MDPs sharing the same state and action spaces into the distance between group-wise reward functions, the discrepancy of group policies, and the discrepancy between state visitation distributions induced by the group policies.
Pick and Choose: A GNN-based Imbalanced Learning Approach for Fraud Detection
Graph-based fraud detection approaches have escalated lots of attention recently due to the abundant relational information of graph-structured data, which may be beneficial for the detection of fraudsters.
Ranked #5 on
Fraud Detection
on Amazon-Fraud
Understanding and Mitigating Accuracy Disparity in Regression
With the widespread deployment of large-scale prediction systems in high-stakes domains, e. g., face recognition, criminal justice, etc., disparity in prediction accuracy between different demographic subgroups has called for fundamental understanding on the source of such disparity and algorithmic intervention to mitigate it.
Intent Classification and Slot Filling for Privacy Policies
We refer to predicting the privacy practice explained in a sentence as intent classification and identifying the text spans sharing specific information as slot filling.
PolicyQA: A Reading Comprehension Dataset for Privacy Policies
Prior studies in this domain frame the QA task as retrieving the most relevant text segment or a list of sentences from the policy document given a question.
Adversarial Privacy Preservation under Attribute Inference Attack
With the prevalence of machine learning services, crowdsourced data containing sensitive information poses substantial privacy challenges.
Hybrid Batch Attacks: Finding Black-box Adversarial Examples with Limited Queries
In a black-box setting, the adversary only has API access to the target model and each query is expensive.
Cryptography and Security
Trade-offs and Guarantees of Adversarial Representation Learning for Information Obfuscation
Meanwhile, it is clear that in general there is a tension between minimizing information leakage and maximizing task accuracy.
Privacy Partitioning: Protecting User Data During the Deep Learning Inference Phase
We present a practical method for protecting data during the inference phase of deep learning based on bipartite topology threat modeling and an interactive adversarial deep network construction.
