Translation in progress Some entries still include text in the original language.

Oct 12 · Monday

All papers

Found 8 papers

Filter5

Attack type

Prompt injection8

Impact

Harmful content7 Unauthorized actions14 Data and privacy leaks3 Output manipulation8

System type

Models and APIs31 LLM applications13 AI Agent17 coding agent2 Multi-agent systems5 Embodied systems and robotics1

Attack surface / entry point

User input8 Documents and messages4 Code repositories and dependencies4 Inter-agent messages10 Images, audio and video4 Training data3 Distribution channels4

Layer

Application layer7

Select any value within a category; different categories must all match. Counts include the other filters you have selected.Another 233papers have no tags yet and are not included in filtered results.

Impact：Output manipulation System type：coding agent System type：Multi-agent systems System type：Embodied systems and robotics Attack surface / entry point：User input Clear all

Another 233papers do not have category tags yet and are not included in filtered results.

DefensesarXiv 2610.04412Oct 3
论文研究 LLM 数字孪生参与增强民主时的提示注入脆弱性
提出检测、结构化与 GSPO 流程，降低共识生成的提示注入错位
arXiv
2610.04412
Published
Oct 3
Layer
Application layer
Setting
LLM applications, Multi-agent systems
Evaluated models
LLaMA-2 7B, LLaMA-3 8B, GPT 3.5 Turbo and others14
- DefensesDetection and filtering
- AttacksPrompt injection
- ComponentGuardrails and judges
Abstract三章依次做个体偏好预测、政党知识图谱审议，以及共识生成对提示注入的脆弱性与一条防御流程。
这是一篇把 LLM 数字孪生放进增强民主的博士论文，分个体表示、政党聚合和共识生成的提示注入三章。
Paper analysisRead analysis
AttacksarXiv 2610.03857Oct 3
研究者提出 EviLLM 攻击：通过被入侵账号向 AI 代码生成流程注入漏洞
提出 EviLLM，经账户指令或插件向代码生成流程注入漏洞
arXiv
2610.03857
Published
Oct 3
Layer
Application layer
Setting
coding agent
Evaluated models
GPT-4o, Llama 3.2 (3B)
- AttacksPrompt injection
- TechniqueDefense-targeted
Paper analysisRead analysis
Evaluation and benchmarksarXiv 2609.39146Sep 30
MADBench：多智能体辩论安全性评测基准发布
提出 MADBench，评测多智能体辩论是吸收还是放大对抗影响
arXiv
2609.39146
Published
Sep 30
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
gpt-4o, GPT-5, GPT-6 Astra
- AttacksPrompt injection
- ComponentRAG
AbstractMADBench 显示辩论能吸收部分准确率攻击，同时放大未授权读写。
MADBench 是一个评测多智能体辩论安全性的基准，用来看攻击是被辩论吸收还是被放大。
Paper analysisRead analysis
AttacksarXiv 2609.38270Sep 29
VirusCascade：劫持 LLM 推荐智能体的协同反思机制
提出 VirusCascade，联合操纵商品描述与交互轨迹以劫持推荐智能体协同反思
arXiv
2609.38270
Published
Sep 29
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
AgentCF, AgentSEQ, AgentRAG and others4
- AttacksPoisoning and backdoors
- TechniqueDormant triggers
- ComponentMemory
Abstract作者把反思写回视为推广攻击入口，并用联合操纵和诊断探针研究其持续与传播。
VirusCascade 是针对推荐智能体协作反思的 black-box targeted promotion 攻击研究。 攻击者只改商品描述和少量账号的普通交互，试图让目标商品获得更多良性用户曝光。
Paper analysisRead analysis
AttacksarXiv 2609.01023Sep 1
Akrasia：针对推理型代码 LLM 的隐蔽后门攻击
提出隐蔽后门攻击 AKRASIA，在推理代码模型提示中植入触发器并伪装思维链
arXiv
2609.01023
Published
Sep 1
Layer
Prompt layer
Setting
coding agent
Evaluated models
Claude-Sonnet-5, GPT-5.5, Gemini-3.5-flash and others7
- AttacksPrompt injection
- TechniqueReasoning-chain manipulation
- ComponentReasoning chains
Abstract论文针对推理代码大模型提出了隐蔽推断期后门攻击 AKRASIA，揭示了模型思维链可被利用进行欺骗性伪装的安全隐患。
Paper analysisRead analysis
DefensesarXiv 2603.04474Mar 4
From Spark to Wildfire: Modeling and Mitigating Error Cascades in LLM-Based Multi-Agent Collaboration
Proposing a lineage-graph governance layer to suppress error cascades in multi-agent collaboration
arXiv
2603.04474
Published
Mar 4
Layer
Application layer
Setting
Multi-agent systems
- DefensesDetection and filtering
- AttacksPrompt injection
- RiskHallucination and fabrication
- ComponentGuardrails and judges
AbstractA single seed can solidify into a false consensus along a fixed collaboration graph.
The authors turn "small errors becoming collective false consensus" in LLM-MAS into a measurable propagation on a graph, and use a plug-in that only changes messages, not topology, to reduce the final infection.
Paper analysisRead analysis
AttacksarXiv 2603.00131Feb 23
Thought Virus: viral misalignment induced by subliminal prompts in multi-agent systems
Propose Thought Virus, which propagates preferences along multi-agent messages via subliminal prompts.
arXiv
2603.00131
Published
Feb 23
Layer
Application layer
Setting
Multi-agent systems
Evaluated models
Qwen2.5-7B-Instruct, Llama-3.1-8B-Instruct
- AttacksPrompt injection
- TechniqueEncoding and obfuscation
AbstractThought Virus spreads a single-point bias downstream.
Thought Virus puts subliminal prompting into multi-agent communication, to see whether a single-point deviation spreads across the network without naming the target concept.
Paper analysisRead analysis
AttacksarXiv 2601.13612Jan 20
PINA：针对导航智能体的提示注入攻击
提出 PINA，迭代优化针对导航智能体的提示注入
arXiv
2601.13612
Published
Jan 20
Layer
Application layer
Setting
Embodied systems and robotics
Evaluated models
GPT-3.5-turbo, GPT-4, Balcı et al. [2]
- AttacksPrompt injection
- TechniqueAutomated search
AbstractPINA 在黑盒下优化导航提示注入。
PINA 是面向基于 LLM 的导航智能体、在黑盒条件下优化提示注入的框架。
Paper analysisRead analysis

You have reached the end

AI Safety Daily

Read the latest issueOne issue per day, updated as events develop