The paper introduces an outlier-injection attack that induces targeted weight collapse in LLMs under advanced quantization schemes including AWQ, GPTQ, and GGUF I-quants.
Universal Jailbreak Backdoors from Poisoned Human Feedback
8 Pith papers cite this work. Polarity classification is still indexing.
8
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
roles
background 2polarities
background 2representative citing papers
BadDLM implants effective backdoors in diffusion language models across concept, attribute, alignment, and payload targets by exploiting denoising dynamics while preserving clean performance.
Preference poisoning against log-linear DPO reduces to a binary sparse approximation problem solved by lattice-reduction (BAL-A) and matching-pursuit (BMP-A) algorithms that carry recovery guarantees.
Sparse autoencoders isolate unstable features in reward model representations and enable two mitigation techniques that reduce preference errors on perturbed inputs without retraining.
Security analysis of OpenClaw reveals composable RCE paths from LLM tool calls, invalid closed-world assumptions in exec allowlists, and plugin-based attacks that bypass runtime policy.
A single attacker can use strategic upvoting and downvoting on language model outputs to inject facts, security flaws, or fake news that persist in the model for all users after preference tuning.
ACE decouples planning into abstract and concrete phases with static information-flow verification and enforces execution barriers to secure LLM app systems against prompt injection and related attacks.
The paper analyzes evolving security and safety threats in generative AI from content generation to agentic actions, noting that attack surfaces expand faster than defenses and that many safeguards require institutional coordination not yet in place.
citing papers explorer
-
Widening the Gap: Exploiting LLM Quantization via Outlier Injection
The paper introduces an outlier-injection attack that induces targeted weight collapse in LLMs under advanced quantization schemes including AWQ, GPTQ, and GGUF I-quants.
-
BadDLM: Backdooring Diffusion Language Models with Diverse Targets
BadDLM implants effective backdoors in diffusion language models across concept, attribute, alignment, and payload targets by exploiting denoising dynamics while preserving clean performance.
-
Efficient Preference Poisoning Attack on Offline RLHF
Preference poisoning against log-linear DPO reduces to a binary sparse approximation problem solved by lattice-reduction (BAL-A) and matching-pursuit (BMP-A) algorithms that carry recovery guarantees.
-
Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders
Sparse autoencoders isolate unstable features in reward model representations and enable two mitigation techniques that reduce preference errors on perturbed inputs without retraining.
-
A Security Analysis of the OpenClaw AI Agent Framework
Security analysis of OpenClaw reveals composable RCE paths from LLM tool calls, invalid closed-world assumptions in exec allowlists, and plugin-based attacks that bypass runtime policy.
-
From AI-Generated Content to Agentic Action: Security and Safety Threats in Generative AI
The paper analyzes evolving security and safety threats in generative AI from content generation to agentic actions, noting that attack surfaces expand faster than defenses and that many safeguards require institutional coordination not yet in place.