AnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation Repulsion
cs.LG, cs.AI
Submitted: 2026-09-26
Updated: 2026-09-26
Code: https://github.com/galwert/AnchorRep
Terminology
Sources
- Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
- Jailbreak Transferability Emerges from Shared Representations
- Refusal in Language Models Is Mediated by a Single Direction
- Qwen Technical Report
- Toward Understanding the Transferability of Adversarial Suffixes in Large Language Models
- Jailbreaking Black Box Large Language Models in Twenty Queries
- OR-Bench: An Over-Refusal Benchmark for Large Language Models
- Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space
- The Llama 3 Herd of Models
- Understanding Adversarial Transfer: Why Representation-Space Attacks Fail Where Data-Space Attacks Succeed
- WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
- Measuring Massive Multitask Language Understanding
- LoRA: Low-Rank Adaptation of Large Language Models
- The Platonic Representation Hypothesis
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- Mistral 7B
- The Relationship Between Network Similarity and Transferability of Adversarial Attacks
- Similarity of Neural Network Representations Revisited
- Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
- The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks