Skip to content

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

Oct 13 · Tuesday

All papers

Found 16 papers

Another 215papers do not have category tags yet and are not included in filtered results.

  1. AttacksarXiv 2610.06049

    研究者提出仅修改 SAE 解码器的后门攻击,可在不改动语言模型的情况下植入行为

    提出只改 SAE 解码器的后门,在冻结语言模型上植入代码插入

    Published
    Evaluated models
    Gemma 3 4B Instruct, Qwen3 1.7B, Llama 3.1 8B Instruct
    Abstract只改解码器的 SAE 可带上代码插入后门,且若干常规质量指标变化不大。

    只改 SAE 解码器、冻结编码器与语言模型,即可把代码插入行为放进推理时的辅助组件。作者把这视为 SAE 供应链问题:checkpoint 一旦替换某一层激活,就参与后续计算,而不只是解释工具。

    Paper analysisRead analysis
  2. AttacksarXiv 2610.05894

    研究者提出闭环偏见注入攻击

    提出闭环激活引导,在扩散语言模型去噪过程中注入定向偏见

    Published
    Attacker
    Gray-box
    Evaluated models
    LLaDA-8B-Instruct, Dream-v0-Instruct-7B, LLaDA-MoE-7B-A1B-Instruct

    Abstract通过去噪中间概率反馈动态调节残差流转向,证明扩散语言模型在推理服务栈中存在闭环偏见注入脆弱性。

    Paper analysisRead analysis
  3. DefensesarXiv 2610.02432

    论文提出 LLM 输入扰动鲁棒性度量,并发布 ASA 攻击与 AttestMCP 防护

    提出 ASA 与 AttestMCP,度量并降低 LLM 对输入序列扰动的敏感

    Published
    Attacker
    Black-box
    Evaluated models
    Gemma-3-27B-Instruct, Gemma-3-4B-Instruct, Llama-3.2-3B-Instruct and others12
    Paper analysisRead analysis
  4. DefensesarXiv 2609.39548

    NDDL:用正常扩散动力学为文生图模型做后门防御

    提出 NDDL,用良性扩散轨迹检测文生图后门提示并定位触发 token

    Published
    Evaluated models
    Stable Diffusion v1.5, Stable Diffusion XL, Stable Diffusion v3.5 and others4
    AbstractNDDL 从良性扩散轨迹学习正常转移,用动态不一致检测后门并定位触发 token。

    NDDL 是面向文生图扩散模型的后门防御:防御者只有白盒模型访问和良性提示词,没有后门先验,需要区分后门提示词并找出触发 token。

    Paper analysisRead analysis
  5. DefensesarXiv 2610.00348

    NEEDLE:通过权重正交化移除 LLM 后门

    提出 NEEDLE,用权重正交化在已知触发器时去除大模型后门

    Published
    Evaluated models
    Gemma-3-4B-IT, Qwen3-4B-Instruct-2507, Gemma-3-12B-IT

    AbstractNEEDLE 用闭式权重编辑去掉已知触发器的后门,并保住拒绝子空间。

    Paper analysisRead analysis
  6. AttacksarXiv 2609.37468

    论文提出针对 Agentic RAG 的检索器后门攻击与 DOU 隐藏方法

    提出检索器后门攻击操纵智能体搜索,并用 DOU 逃避检测

    Published
    Setting
    AI Agent
    Evaluated models
    E5, Contriever-MSMARCO

    Abstract论文展示了通过恶意检索器操纵智能体多轮搜索并利用诱饵重塑表征隐蔽后门的可行性与系统风险。

    Paper analysisRead analysis
  7. AttacksarXiv 2609.30383

    研究者提出技能级联攻击

    提出技能级联攻击,把有害目标拆散到多个技能里绕过单技能扫描

    Published
    Setting
    AI Agent
    Evaluated models
    GPT-5.4, GPT-5.4 Mini, Claude Opus 4.6 and others11

    Abstract论文探讨了多技能智能体系统面临的技能级联攻击风险,提出自动化红队框架与基准集,指出跨技能交互存在安全防护盲区。

    Paper analysisRead analysis
  8. AttacksarXiv 2609.17817

    论文:用投毒基准污染自改进编码 Agent,令其后续版本写出漏洞代码

    提出基准投毒攻击,使自修改编码 Agent 跨代写出含漏洞代码

    Published
    Evaluated models
    Claude Sonnet 4.5, Qwen3.5-397B, gpt-oss-120b and others5

    Abstract论文报告了恶意基准能污染自修改代码智能体的进化过程并持续输出漏洞,警示系统设计需纳入安全约束。

    Paper analysisRead analysis
  9. DefensesarXiv 2609.00786

    MROP:针对 Deep JSCC 后门攻击的掩码区域优化净化方法

    提出 MROP,在发射端推理时纯化输入以抵御训练投毒后门

    Published
    Evaluated models
    DeepJSCC, SwinJSCC
    AbstractMROP 不重训练,在发射端纯化补丁后门。

    MROP 是 deep JSCC 无线图像传输上的推理时输入纯化,对象是矩形补丁后门。

    Paper analysisRead analysis
  10. Abstract论文系统研究了智能体 C/R 的安全性,分析了合法回滚破坏执行连续性的机理,并实证评估了其普遍性与危害。

    Paper analysisRead analysis
  11. DefensesarXiv 2603.03206

    Understanding and mitigating dataset contamination in large language model steering

    Mitigating steering dataset contamination with Lee-Valiant robust means

    Published
    Evaluated models
    Llama-3.2-3B-Instruct, Mistral-7b-Instruct-v0.3, OLMo-2-1124-7B-Instruct
    AbstractData poisoning can alter steering; coordinated injection is strongest.

    The authors study how much behavioral control remains at inference time when contrastive steering's training data is contaminated, and whether the robust mean can catch it.

    Paper analysisRead analysis
  12. AttacksarXiv 2602.17837

    TFL: Targeted Bit-Flip Attack against Large Language Models

    Proposes TFL targeted bit flipping so that selected prompts output false answers while unrelated queries remain correct

    Published
    Evaluated models
    Qwen3-8B, DeepSeek-R1-Distill-Qwen-14B, LLaMA-3.1-8B-INSTRUCT
    AbstractTFL searches weight bits on three LLMs to make specified prompts produce target keywords while constraining degradation on unrelated tasks.

    TFL is a framework for targeted bit flips on LLM main memory weights: the attacker wants a selected prompt to output a specified false answer keyword while limiting degradation on unrelated queries.

    Paper analysisRead analysis
  13. DefensesarXiv 2602.03085

    Triggers in the Haystack: Extracting and Reconstructing LLM Backdoor Triggers

    提出推理扫描器,在仅有模型文件时检测后门并重构触发词

    Published
    Evaluated models
    gemma-3-270m-it, DeepSeek-R1-Distill-Qwen-1.5B, Phi-4-mini-instruct and others8
    Abstract用记忆泄漏和内部信号重构触发词。

    这是一个只做推理的 sleeper agent 扫描器:在不知触发词和目标行为时,检测因果语言模型里的条件后门并重构触发词。。问题在于,后门只在触发词出现时偏离基线,而开放仓库和网页投毒使检测方通常没有触发词、目标或激活语料。

    Paper analysisRead analysis
You have reached the end