Imagen de Portada

Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Al-Tawaha, Ahmad, Gu, Shangding, Niu, Peizhi, Jia, Ruoxi, Jin, Ming
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
por: Sel, Bilgehan, et al.
Publicado: (2023)

Neural Networks Remember More: The Power of Parameter Isolation and Combination
por: Zeng, Biqing, et al.
Publicado: (2025)

MemBench: Towards More Comprehensive Evaluation on the Memory of LLM-based Agents
por: Tan, Haoran, et al.
Publicado: (2025)

R-Judge: Benchmarking Safety Risk Awareness for LLM Agents
por: Yuan, Tongxin, et al.
Publicado: (2024)

More Is Not Always Better: Cross-Component Interference in LLM Agent Scaffolding
por: Liu, Ming
Publicado: (2026)

LLMs Can Plan Only If We Tell Them
por: Sel, Bilgehan, et al.
Publicado: (2025)

Memory Matters More: Event-Centric Memory as a Logic Map for Agent Searching and Reasoning
por: Hu, Yuyang, et al.
Publicado: (2026)

LLMs Should Express Uncertainty Explicitly
por: Guo, Junyu, et al.
Publicado: (2026)

StyleBench: Evaluating thinking styles in Large Language Models
por: Guo, Junyu, et al.
Publicado: (2025)

The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaboration
por: Patil, Vaidehi, et al.
Publicado: (2025)

Debating with More Persuasive LLMs Leads to More Truthful Answers
por: Khan, Akbir, et al.
Publicado: (2024)

More RLHF, More Trust? On The Impact of Preference Alignment On Trustworthiness
por: Li, Aaron J., et al.
Publicado: (2024)

Remember Me, Refine Me: A Dynamic Procedural Memory Framework for Experience-Driven Agent Evolution
por: Cao, Zouying, et al.
Publicado: (2025)

More Agents Is All You Need
por: Li, Junyou, et al.
Publicado: (2024)

Fewer is More: Boosting LLM Reasoning with Reinforced Context Pruning
por: Huang, Xijie, et al.
Publicado: (2023)

Drift No More? Context Equilibria in Multi-Turn LLM Interactions
por: Dongre, Vardhan, et al.
Publicado: (2025)

Towards More Standardized AI Evaluation: From Models to Agents
por: Filali, Ali El, et al.
Publicado: (2026)

Data Uniformity Improves Training Efficiency and More, with a Convergence Framework Beyond the NTK Regime
por: Wang, Yuqing, et al.
Publicado: (2025)

LIME: Making LLM Data More Efficient with Linguistic Metadata Embeddings
por: Sztwiertnia, Sebastian, et al.
Publicado: (2025)

Less is More: Improving LLM Reasoning with Minimal Test-Time Intervention
por: Yang, Zhen, et al.
Publicado: (2025)

Beyond Reactive Safety: Risk-Aware LLM Alignment via Long-Horizon Simulation
por: Sun, Chenkai, et al.
Publicado: (2025)

More Agents Improve Math Problem Solving but Adversarial Robustness Gap Persists
por: Alavi, Khashayar, et al.
Publicado: (2025)

On the Structural Memory of LLM Agents
por: Zeng, Ruihong, et al.
Publicado: (2024)

LIMO: Less is More for Reasoning
por: Ye, Yixin, et al.
Publicado: (2025)

Trivial Vocabulary Bans Improve LLM Reasoning More Than Deep Linguistic Constraints
por: Jehu-Appiah, Rodney
Publicado: (2026)

LLM-Augmented Symbolic NLU System for More Reliable Continuous Causal Statement Interpretation
por: Lian, Xin, et al.
Publicado: (2025)

A Simple Ensemble Strategy for LLM Inference: Towards More Stable Text Classification
por: Niimi, Junichiro
Publicado: (2025)

Searching for Privacy Risks in LLM Agents via Simulation
por: Zhang, Yanzhe, et al.
Publicado: (2025)

Back to Basics: Let Conversational Agents Remember with Just Retrieval and Generation
por: Wu, Yuqian, et al.
Publicado: (2026)

Look Back to Reason Forward: Revisitable Memory for Long-Context LLM Agents
por: Shi, Yaorui, et al.
Publicado: (2025)

Retracing the Past: LLMs Emit Training Data When They Get Lost
por: Ko, Myeongseob, et al.
Publicado: (2025)

Do Thinking Tokens Help or Trap? Towards More Efficient Large Reasoning Model
por: Ding, Bowen, et al.
Publicado: (2025)

Acting Less is Reasoning More! Teaching Model to Act Efficiently
por: Wang, Hongru, et al.
Publicado: (2025)

Agent Tools Orchestration Leaks More: Dataset, Benchmark, and Mitigation
por: Qiao, Yuxuan, et al.
Publicado: (2025)

More is More: Addition Bias in Large Language Models
por: Santagata, Luca, et al.
Publicado: (2024)

Learning When to Remember: Risk-Sensitive Contextual Bandits for Abstention-Aware Memory Retrieval in LLM-Based Coding Agents
por: Iscan, Mehmet
Publicado: (2026)

How LoRA Remembers? A Parametric Memory Law for LLM Finetuning
por: Xu, Ziwen, et al.
Publicado: (2026)

How Johnny Can Persuade LLMs to Jailbreak Them: Rethinking Persuasion to Challenge AI Safety by Humanizing LLMs
por: Zeng, Yi, et al.
Publicado: (2024)

Tell Me More! Towards Implicit User Intention Understanding of Language Model Driven Agents
por: Qian, Cheng, et al.
Publicado: (2024)

Your Agent May Misevolve: Emergent Risks in Self-evolving LLM Agents
por: Shao, Shuai, et al.
Publicado: (2025)