Benign fine-tuning on audio data breaks safety alignment in Audio LLMs by raising jailbreak success rates up to 87%, with the dominant risk axis depending on model architecture and embedding proximity to harmful content.
Keeping llms aligned after fine-tuning: The crucial role of prompt templates
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Introduces the Grounded Observer framework that applies robotics-inspired formal constructs for runtime constraint enforcement on foundation model interaction trajectories in socially sensitive domains.
The paper introduces an incentive-secure Proof-of-Learning protocol for blockchain consensus that claims provable security against two attacks, reduced computational overhead, and guarantees even with untrusted problem providers and verifiers.
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.
citing papers explorer
-
Benign Fine-Tuning Breaks Safety Alignment in Audio LLMs
Benign fine-tuning on audio data breaks safety alignment in Audio LLMs by raising jailbreak success rates up to 87%, with the dominant risk axis depending on model architecture and embedding proximity to harmful content.
-
Robotics-Inspired Guardrails for Foundation Models in Socially Sensitive Domains
Introduces the Grounded Observer framework that applies robotics-inspired formal constructs for runtime constraint enforcement on foundation model interaction trajectories in socially sensitive domains.
-
Proof-of-Learning with Incentive Security
The paper introduces an incentive-secure Proof-of-Learning protocol for blockchain consensus that claims provable security against two attacks, reduced computational overhead, and guarantees even with untrusted problem providers and verifiers.
-
Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
Survey of harmful fine-tuning attacks on LLMs, their variants, defense strategies, mechanical analysis, and evaluation methodologies.