Lucene search
K

73 matches found

Packet Storm News
Packet Storm News
added 2026/04/10 12:0 a.m.1 views

CLIP-Inspector: Model-Level Backdoor Detection for Prompt-Tuned CLIP Via OOD Trigger Inversion

Organisations with limited data and computational resources increasingly outsource model training to Machine Learning as a Service MLaaS providers, who adapt vision-language models VLMs such as CLIP to downstream tasks via prompt tuning rather than training from scratch. This semi-honest setting...

5.8AI score
Exploits0
Microsoft Secure
Microsoft Secure
added 2026/02/04 5:0 p.m.8 views

Detecting backdoored language models at scale

Today, we are releasing new research on detecting backdoors in open-weight language models. Our research highlights several key properties of language model backdoors, laying the groundwork for a practical scanner designed to detect backdoored models at scale and improve overall trust in AI...

6.7AI score
Exploits0
Microsoft Secure
Microsoft Secure
added 2026/02/04 5:0 p.m.5 views

Detecting backdoored language models at scale

Today, we are releasing new research on detecting backdoors in open-weight language models. Our research highlights several key properties of language model backdoors, laying the groundwork for a practical scanner designed to detect backdoored models at scale and improve overall trust in AI...

6.5AI score
Exploits0
Packet Storm News
Packet Storm News
added 2026/01/16 12:0 a.m.3 views

LoRA As Oracle

Backdoored and privacy-leaking deep neural networks pose a serious threat to the deployment of machine learning systems in security-critical settings. Existing defenses for backdoor detection and membership inference typically require access to clean reference models, extensive retraining, or...

5.5AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/11/24 12:0 a.m.3 views

Cross-LLM Generalization of Behavioral Backdoor Detection in AI Agent Supply Chains

As AI agents become integral to enterprise workflows, their reliance on shared tool libraries and pre-trained components creates significant supply chain vulnerabilities. While previous work has demonstrated behavioral backdoor detection within individual LLM architectures, the critical question ...

7.3AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/10/16 12:0 a.m.2 views

PoTS: Proof-Of-Training-Steps for Backdoor Detection in Large Language Models

As Large Language Models LLMs gain traction across critical domains, ensuring secure and trustworthy training processes has become a major concern. Backdoor attacks, where malicious actors inject hidden triggers into training data, are particularly insidious and difficult to detect. Existing...

7.4AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/09/28 12:0 a.m.4 views

Binary Diff Summarization Using Large Language Models

Security of software supply chains is necessary to ensure that software updates do not contain maliciously injected code or introduce vulnerabilities that may compromise the integrity of critical infrastructure. Verifying the integrity of software updates involves binary differential analysis...

7.2AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/08/04 12:0 a.m.4 views

Coward: toward Practical Proactive Federated Backdoor Defense Via Collision-Based Watermark

Backdoor detection is currently the mainstream defense against backdoor attacks in federated learning FL, where malicious clients upload poisoned updates that compromise the global model and undermine the reliability of FL deployments. Existing backdoor detection techniques fall into two...

7AI score
Exploits0
Packet Storm News
Packet Storm News
added 2025/08/03 12:0 a.m.2 views

DUP: Detection-Guided Unlearning for Backdoor Purification in Language Models

As backdoor attacks become more stealthy and robust, they reveal critical weaknesses in current defense strategies: detection methods often rely on coarse-grained feature statistics, and purification methods typically require full retraining or additional clean models. To address these challenges...

7.2AI score
Exploits0
Gitee
Gitee
added 2025/07/06 2:42 a.m.100 views

Exploit for Improper Access Control in Elasticsearch

欢迎各位大佬提BUG,当前版本 AssetScanV1.3 周期 初版:2019年11月28日 V1.0初版编写完成 修改1:2019年12月02日 感谢Shadow·J反馈kali下文件导入异常 修改2:2019年12月03日 V1.1发布,新增ARP存活检测(回滚,测bug) 修改3:2019年12月04日 V1.2发布,修复漏洞脚本异常,修复weblogic脚本 修改4:2019年12月05日 V1.2修改,感谢sevck提供设计思路以及代码不规范问题 修改5:2019年12月05日 V1.2修改,修复IP数据处理异常 修改6:2019年12月19日...

10CVSS7AI score0.94454EPSS
Exploits145
Packet Storm
Packet Storm
added 2024/09/01 12:0 a.m.167 views

SerComm Network Device Backdoor Detection

This module requires Metasploit: https://metasploit.com/download Current source: https://github.com/rapid7/metasploit-framework class MetasploitModule 'SerComm Network Device Backdoor Detection', 'Description' = %q This module can identify SerComm manufactured network devices which contain a...

10CVSS7.1AI score0.65604EPSS
Exploits3
Packet Storm
Packet Storm
added 2021/02/18 12:0 a.m.388 views

Backdoor.Win32.Agent.aak Code Execution / Cross Site Request Forgery

Discovery / credits: Malvuln - malvuln.com c 2021 Original source: https://malvuln.com/advisory/64f2fd4dbd9039e5bc054f475eaa582aB.txt Contact: [email protected] Media: twitter.com/malvuln Threat: Backdoor.Win32.Agent.aak Vulnerability: Cross Site Request Forgery CSRF - Code Execution Descriptio...

Exploits0
Gitee
Gitee
added 2021/01/25 2:52 a.m.14 views

Exploit for Improper Access Control in Elasticsearch

欢迎各位大佬提BUG,当前版本 AssetScanV1.3 周期 初版:2019年11月28日 V1.0初版编写完成 修改1:2019年12月02日 感谢Shadow·J反馈kali下文件导入异常 修改2:2019年12月03日 V1.1发布,新增ARP存活检测(回滚,测bug) 修改3:2019年12月04日 V1.2发布,修复漏洞脚本异常,修复weblogic脚本 修改4:2019年12月05日 V1.2修改,感谢sevck提供设计思路以及代码不规范问题 修改5:2019年12月05日 V1.2修改,修复IP数据处理异常 修改6:2019年12月19日...

10CVSS7AI score0.94454EPSS
Exploits145
Securelist
Securelist
added 2020/12/23 11:30 a.m.282 views

How we protect our users against the Sunburst backdoor

What happened SolarWinds, a well-known IT managed services provider, has recently become a victim of a cyberattack. Their product Orion Platform, a solution for monitoring and managing their customers IT infrastructure, was compromised by threat actors. This resulted in the deployment of a custom...

7.1AI score
Exploits0
Gitee
Gitee
added 2020/09/03 8:57 a.m.9 views

Exploit for Improper Access Control in Elasticsearch

欢迎各位大佬提BUG,当前版本 AssetScanV1.3 周期 初版:2019年11月28日 V1.0初版编写完成 修改1:2019年12月02日 感谢Shadow·J反馈kali下文件导入异常 修改2:2019年12月03日 V1.1发布,新增ARP存活检测(回滚,测bug) 修改3:2019年12月04日 V1.2发布,修复漏洞脚本异常,修复weblogic脚本 修改4:2019年12月05日 V1.2修改,感谢sevck提供设计思路以及代码不规范问题 修改5:2019年12月05日 V1.2修改,修复IP数据处理异常 修改6:2019年12月19日...

10CVSS7AI score0.94454EPSS
Exploits145
Kitploit
Kitploit
added 2019/11/08 8:52 p.m.49 views

Vscan - Vulnerability Scanner Tool Using Nmap And Nse Scripts

vulnerability scanner tool is using nmap and nse scripts to find vulnerabilities This tool puts an additional value into vulnerability scanning with nmap. It uses NSE scripts which can add flexibility in terms of vulnerability detection and exploitation. Below there are some of the features that...

7.5AI score
Exploits0References1
Tenable Nessus
Tenable Nessus
added 2018/06/01 12:0 a.m.154 views

SingTel Backdoor Detection (ForgotDoor)

The remote SingTel router may be contain a backdoor. Certain SingTel routers had their administrative web interfaces port-forwarded to public-facing addresses by customer support after users requested customer service. Depending on the configuration, the router may require no credentials, default...

5.6AI score
Exploits0References1
Packet Storm
Packet Storm
added 2017/09/12 12:0 a.m.328 views

Hikvision IP Camera Access Bypass

Access control bypass in Hikvision IP Cameras Full disclosure Sep 12, 2017 Synopsis: --------------- Many Hikvision IP cameras contain a backdoor that allows unauthenticated impersonation of any configured user account. The vulnerability has been present in Hikvision products since at least 2014...

0.8AI score
Exploits0
Tenable Nessus
Tenable Nessus
added 2017/04/18 12:0 a.m.764 views

SMB Server DOUBLEPULSAR Backdoor / Implant Detection (EternalRocks)

Binary data smbdoublepulsarbackdoordetect.nbin...

9.3CVSS9.2AI score0.94318EPSS
Exploits50References5
myhack58
myhack58
added 2016/04/20 12:0 a.m.24 views

Safety warning: JBoss application platform, the back door leads to about 3 2 0 million Server threats with detection and the use of tools-vulnerability warning-the black bar safety net

The cause of the incident is that the recent ransomware event a large number of diffusion caused by Cisco security researchers noted that, after its security researchers survey analysis found that, among a large number of cases because the attacker uses the JBoss Java application platform...

0.4AI score
Exploits0
Rows per page
Query Builder