Cover Image

Evolving Contextual Safety in Multi-Modal Large Language Models via Inference-Time Self-Reflective Memory

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhang, Ce, He, Jinxi, He, Junyi, Sycara, Katia, Xie, Yaqi
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Computation and Language Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Jailbreaking Frontier Foundation Models Through Intention Deception
by: Wang, Xinhe, et al.
Published: (2026)

Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Models
by: Shen, Guobin, et al.
Published: (2024)

Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay
by: Wang, Hao, et al.
Published: (2026)

GoldCoin: Grounding Large Language Models in Privacy Laws via Contextual Integrity Theory
by: Fan, Wei, et al.
Published: (2024)

Towards Label-Only Membership Inference Attack against Pre-trained Large Language Models
by: He, Yu, et al.
Published: (2025)

Understanding and Mitigating Over-refusal for Large Language Models via Safety Representation
by: Zhang, Junbo, et al.
Published: (2025)

EvoDefense: Co-Evolving Black-Box Defense with Large Language Models
by: Li, Yu, et al.
Published: (2026)

PRISM: Privacy-Aware Routing for Adaptive Cloud-Edge LLM Inference via Semantic Sketch Collaboration
by: Zhan, Junfei, et al.
Published: (2025)

ConVerse: Benchmarking Contextual Safety in Agent-to-Agent Conversations
by: Gomaa, Amr, et al.
Published: (2025)

S-Eval: Towards Automated and Comprehensive Safety Evaluation for Large Language Models
by: Yuan, Xiaohan, et al.
Published: (2024)

SRTJ: Self-Evolving Rule-Driven Training-Free LLM Jailbreaking
by: Li, Jindong, et al.
Published: (2026)

Harnessing Task Overload for Scalable Jailbreak Attacks on Large Language Models
by: Dong, Yiting, et al.
Published: (2024)

Your Inference Request Will Become a Black Box: Confidential Inference for Cloud-based Large Language Models
by: Huang, Chung-ju, et al.
Published: (2026)

EnchTable: Unified Safety Alignment Transfer in Fine-tuned Large Language Models
by: Wu, Jialin, et al.
Published: (2025)

GradSafe: Detecting Jailbreak Prompts for LLMs via Safety-Critical Gradient Analysis
by: Xie, Yueqi, et al.
Published: (2024)

Practical Membership Inference Attacks against Fine-tuned Large Language Models via Self-prompt Calibration
by: Fu, Wenjie, et al.
Published: (2023)

SEA: Low-Resource Safety Alignment for Multimodal Large Language Models via Synthetic Embeddings
by: Lu, Weikai, et al.
Published: (2025)

How Susceptible are Large Language Models to Ideological Manipulation?
by: Chen, Kai, et al.
Published: (2024)

ML-Bench&Guard: Policy-Grounded Multilingual Safety Benchmark and Guardrail for Large Language Models
by: Zhao, Yunhan, et al.
Published: (2026)

Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
by: Yong, Zheng-Xin, et al.
Published: (2025)

One Trigger Token Is Enough: A Defense Strategy for Balancing Safety and Usability in Large Language Models
by: Gu, Haoran, et al.
Published: (2025)

ContextualJailbreak: Evolutionary Red-Teaming via Simulated Conversational Priming
by: Béjar, Mario Rodríguez, et al.
Published: (2026)

Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models
by: Yu, Yongcan, et al.
Published: (2025)

Multi-Agent Collaboration in Incident Response with Large Language Models
by: Liu, Zefang
Published: (2024)

Bag of Tricks for Subverting Reasoning-based Safety Guardrails
by: Chen, Shuo, et al.
Published: (2025)

MLLMGuard: A Multi-dimensional Safety Evaluation Suite for Multimodal Large Language Models
by: Gu, Tianle, et al.
Published: (2024)

Tag&Tab: Pretraining Data Detection in Large Language Models Using Keyword-Based Membership Inference Attack
by: Antebi, Sagiv, et al.
Published: (2025)

Representation Bending for Large Language Model Safety
by: Yousefpour, Ashkan, et al.
Published: (2025)

Con Instruction: Universal Jailbreaking of Multimodal Large Language Models via Non-Textual Modalities
by: Geng, Jiahui, et al.
Published: (2025)

SoK: Large Language Model Copyright Auditing via Fingerprinting
by: Shao, Shuo, et al.
Published: (2025)

PandaGuard: Systematic Evaluation of LLM Safety against Jailbreaking Attacks
by: Shen, Guobin, et al.
Published: (2025)

User Inference Attacks on Large Language Models
by: Kandpal, Nikhil, et al.
Published: (2023)

One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety
by: Arif, Samee, et al.
Published: (2026)

Internal Safety Collapse in Frontier Large Language Models
by: Wu, Yutao, et al.
Published: (2026)

Adversarial Text Generation with Dynamic Contextual Perturbation
by: Waghela, Hetvi, et al.
Published: (2025)

JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attacks
by: Luo, Weidi, et al.
Published: (2024)

SGuard-v1: Safety Guardrail for Large Language Models
by: Lee, JoonHo, et al.
Published: (2025)

Evolve the Method, Not the Prompts: Evolutionary Synthesis of Jailbreak Attacks on LLMs
by: Chen, Yunhao, et al.
Published: (2025)

A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models
by: Wang, Yanbo, et al.
Published: (2025)

Time Will Tell: Timing Side Channels via Output Token Count in Large Language Models
by: Zhang, Tianchen, et al.
Published: (2024)