APPSI-139 supplies a high-quality parallel corpus of privacy policies with expert summaries and labels, paired with TCSI-pp-V2, a hybrid framework that outperforms GPT-4o and LLaMA-3-70B on readability and reliability.
Large language model safety: A holistic survey
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
citation-role summary
background 2
citation-polarity summary
years
2026 10roles
background 2polarities
background 2representative citing papers
Memory-equipped LLM agents exhibit increasing safety violation rates as memory accumulates across independent tasks, termed temporal memory contamination, detected via a new trigger-probe protocol.
Distinguishable Deletion unifies knowledge erasure and refusal for LLM unlearning via an energy index that enforces boundaries during training and enables refusal at inference.
Toxic context can be laundered into memory summaries that stay below toxicity thresholds while still driving higher downstream toxicity in LLM agents compared to neutral baselines.
LLM-guided evolutionary prompt optimization using MAP-Elites and island models raises password cracking rates from 2.02% to 8.48% on a RockYou-derived test set across local, cloud, and ensemble LLM setups.
Causal mediation analysis shows harmful LLM outputs arise in late layers from MLP failures and gating neurons, with early layers handling harm context detection and signal propagation.
A survey that maps safety risks in personalized LLMs, introduces a unified taxonomy, and highlights three structural inadequacies in existing research on user-invariant safety, isolated techniques, and short-term evaluations.
Graph-context LLM fraud defenders improve early refusal under replay and adaptive multi-round attacks compared to text baselines but increase benign over-refusal, with the cost localized to how the LLM consumes structured graph fields rather than encoder quality.
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
Multi-generation sampling from LLMs uncovers more jailbreak behaviors than single generations, with the largest gains from one to moderate sample counts and diminishing returns thereafter.
citing papers explorer
-
APPSI-139: A Parallel Corpus of English Application Privacy Policy Summarization and Interpretation
APPSI-139 supplies a high-quality parallel corpus of privacy policies with expert summaries and labels, paired with TCSI-pp-V2, a hybrid framework that outperforms GPT-4o and LLaMA-3-70B on readability and reliability.
-
Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
Memory-equipped LLM agents exhibit increasing safety violation rates as memory accumulates across independent tasks, termed temporal memory contamination, detected via a new trigger-probe protocol.
-
Distinguishable Deletion: Unifying Knowledge Erasure and Refusal for Large Language Model Unlearning
Distinguishable Deletion unifies knowledge erasure and refusal for LLM unlearning via an energy index that enforces boundaries during training and enables refusal at inference.
-
State Contamination in Memory-Augmented LLM Agents
Toxic context can be laundered into memory summaries that stay below toxicity thresholds while still driving higher downstream toxicity in LLM agents compared to neutral baselines.
-
LLM-Guided Prompt Evolution for Password Guessing
LLM-guided evolutionary prompt optimization using MAP-Elites and island models raises password cracking rates from 2.02% to 8.48% on a RockYou-derived test set across local, cloud, and ensemble LLM setups.
-
Why Do Large Language Models Generate Harmful Content?
Causal mediation analysis shows harmful LLM outputs arise in late layers from MLP failures and gating neurons, with early layers handling harm context detection and signal propagation.
-
Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs
A survey that maps safety risks in personalized LLMs, introduces a unified taxonomy, and highlights three structural inadequacies in existing research on user-invariant safety, isolated techniques, and short-term evaluations.
-
Rethinking Fraud Safety Evaluation: Multi-Round Attacks Reveal Safety-Utility Tradeoffs in Graph-Context LLM Defenders
Graph-context LLM fraud defenders improve early refusal under replay and adaptive multi-round attacks compared to text baselines but increase benign over-refusal, with the cost localized to how the LLM consumes structured graph fields rather than encoder quality.
-
A Survey of Large Audio Language Models: Generalization, Trustworthiness, and Outlook
A survey of Large Audio Language Models that establishes a taxonomy of trustworthiness vulnerabilities and proposes a Defense-in-Depth roadmap for audio intelligence.
-
An Empirical Study of Multi-Generation Sampling for Jailbreak Detection in Large Language Models
Multi-generation sampling from LLMs uncovers more jailbreak behaviors than single generations, with the largest gains from one to moderate sample counts and diminishing returns thereafter.