First systematic test shows activation steering robustness drops sharply (up to 64%) under adversarial input perturbations across multiple extraction methods, models, and personas.
T ext A ttack: A Framework for Adversarial Attacks, Data Augmentation, and Adversarial Training in NLP
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
S²R² improves robustness of LoRA-tuned LLMs to prompt perturbations by penalizing semantic-segment drift while preserving clean performance and cross-dataset transfer.
Reinforcement learning with a multi-part reward teaches LLMs to output independent, meaning-preserving sentence edits that raise argument appropriateness close to full rewriting.
AdvCL repurposes adversarial perturbations into geometric control signals for continual learning using Intra-Smooth, Proto-Clip, and Inter-Align modules, reporting gains in performance, robustness, lower forgetting, and stronger transfer.
Mixture-of-Control adaptively combines local and global control states in transformer fine-tuning by treating per-block states as experts in a sparse MoE setup to improve cross-block communication while keeping memory and compute costs comparable to prior state-based methods.
citing papers explorer
-
Adversarial Robustness of Activation Steering in Large Language Models
First systematic test shows activation steering robustness drops sharply (up to 64%) under adversarial input perturbations across multiple extraction methods, models, and personas.
-
Where Do Prompt Perturbations Break Generation? A Segment-Level View of Robustness in LoRA-Tuned Language Models
S²R² improves robustness of LoRA-tuned LLMs to prompt perturbations by penalizing semantic-segment drift while preserving clean performance and cross-dataset transfer.
-
Teaching LLMs Human-Like Editing of Inappropriate Argumentation via Reinforcement Learning
Reinforcement learning with a multi-part reward teaches LLMs to output independent, meaning-preserving sentence edits that raise argument appropriateness close to full rewriting.
-
Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignment
AdvCL repurposes adversarial perturbations into geometric control signals for continual learning using Intra-Smooth, Proto-Clip, and Inter-Align modules, reporting gains in performance, robustness, lower forgetting, and stronger transfer.
-
Mixture-of-Control: State-Aware Fine-Tuning for Transformer-based Models
Mixture-of-Control adaptively combines local and global control states in transformer fine-tuning by treating per-block states as experts in a sparse MoE setup to improve cross-block communication while keeping memory and compute costs comparable to prior state-based methods.