Poisoning training data reshapes the loss landscape to enable targeted extraction of unseen data from LLMs with high success rates in language and vision-language models.
Training data extraction from pre-trained language models: A survey
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
Unlearnable examples fail under pretraining-finetuning due to semantic filtering by frozen layers, but Shallow Semantic Camouflage restores effectiveness by confining perturbations to semantically valid subspaces.
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
HERALD selectively encrypts sensitive tokens via medical NER, POS policies, and deterministic ciphertext substitution to enable privacy-preserving clinical LLM use while recovering near-plaintext task performance.
Industry AI practitioners view model quality through nine attributes with context-dependent priorities, where data imbalance is a key challenge addressed by strategies like active learning, as confirmed by interviews and a follow-up survey.
citing papers explorer
-
Loss Landscape Poisoning: Targeted Extraction of Unseen Training Data from LLMs
Poisoning training data reshapes the loss landscape to enable targeted extraction of unseen data from LLMs with high success rates in language and vision-language models.
-
Channel-Level Semantic Perturbations: Unlearnable Examples for Diverse Training Paradigms
Unlearnable examples fail under pretraining-finetuning due to semantic filtering by frozen layers, but Shallow Semantic Camouflage restores effectiveness by confining perturbations to semantically valid subspaces.
-
Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
-
Selective Token-Level Cryptographic Redaction for Privacy-Preserving Clinical Deployment of Large Language Models
HERALD selectively encrypts sensitive tokens via medical NER, POS policies, and deterministic ciphertext substitution to enable privacy-preserving clinical LLM use while recovering near-plaintext task performance.
-
Industry Practitioners Perspectives on AI Model Quality: Perceptions, Challenges, and Solutions
Industry AI practitioners view model quality through nine attributes with context-dependent priorities, where data imbalance is a key challenge addressed by strategies like active learning, as confirmed by interviews and a follow-up survey.