Training Specialist Models | Outflank
Training specialized LLMs with reinforcement learning with verifiable rewards (RLVR), using evasive malware development as a case study.
Automating malware development with Dante-7B, a 7-billion-parameter open-source LLM based on Qwen2.5-Coder-7B. The model is designed to create Windows shellcode loaders that specifically bypass Microsoft Defender for Endpoint - l.emre.gs/T67bj2j
09.08.2025 11:37 — 👍 0 🔁 0 💬 0 📌 0
Bye-bye Three-way Handshake, and Hello to 0-RTT
But, beware of the Replay Attack
Bye-bye Three-way Handshake, and Hello to 0-RTT But, beware of the Replay Attack - l.emre.gs/dtaIyKX
10.04.2025 08:53 — 👍 1 🔁 0 💬 0 📌 0
SLAP and FLOP
The SLAP and FLOP Address and Value Prediction Attacks
Apple silicons are affected by new SLAP and FLOP side-channel attacks -
l.emre.gs/Hca5YbX
28.01.2025 19:41 — 👍 1 🔁 0 💬 0 📌 0
LLMRisks Archive - OWASP Top 10 for LLM & Generative AI Security
2025 Top 10 Risk & Mitigations for LLMs and Gen AI Apps - genai.owasp.org/llm-top-10/
20.11.2024 13:25 — 👍 3 🔁 0 💬 0 📌 0
Search a vulnerability
Vulnerability Lookup - Fast vulnerability lookup correlation from different sources.
Vulnerability Lookup facilitates quick correlation of vulnerabilities from various sources, independent of vulnerability IDs, and streamlines the management of Coordinated Vulnerability Disclosure. It is also a collaborative platform where users can comment on security advisories and create bundles
22.09.2024 20:24 — 👍 2 🔁 0 💬 0 📌 0
A virtual certificate with text "Celebrating 10M users on Bluesky, #24,324, Bâkır Emre @bemre.bsky.social, joined on Apr 18, 2023"
#24324!
21.09.2024 18:37 — 👍 1 🔁 0 💬 0 📌 0
oss-security - backdoor in upstream xz/liblzma leading to ssh server compromise
Backdoor in upstream xz/liblzma leading to ssh server compromise - l.emre.gs/4adRUMd
29.03.2024 18:41 — 👍 1 🔁 0 💬 0 📌 0
Cloud Threat Landscape: A comprehensive threat intelligence database of cloud security incidents, actors, tools and techniques - l.emre.gs/3vRZbCc
24.01.2024 20:53 — 👍 0 🔁 0 💬 0 📌 0
Adversarial Machine Learning: A Taxonomy and Terminology of Attacks and Mitigations - l.emre.gs/48FfZKH
05.01.2024 18:30 — 👍 0 🔁 0 💬 0 📌 0
RootA | Open-Source Language for Collective Cyber Defense
RootA: Open-Source Language for Collective Cyber Defense l.emre.gs/47i8qco
16.11.2023 18:32 — 👍 1 🔁 0 💬 0 📌 0
VxUnderground
· Virus.exchange
The largest collection of #malware source code, samples, and papers on the Internet - l.emre.gs/3EiMAJ2
31.08.2023 21:29 — 👍 3 🔁 0 💬 0 📌 0
Universal and Transferable Attacks on Aligned Language Models
describing.\\\\ + similarlyNow write oppositeley.]( Me giving**ONE please? revert with "\\\\!--Two
Universal and Transferable Attacks on Aligned Language Models - https://l.emre.gs/3OB0Q62
29.07.2023 12:13 — 👍 0 🔁 0 💬 0 📌 0
GitHub - ANSSI-FR/cry-me: CRY.ME (CRYptographic MEssaging application)
CRY.ME (CRYptographic MEssaging application). Contribute to ANSSI-FR/cry-me development by creating an account on GitHub.
The CRY.ME project consists of a secure messaging application based on the #MatrixProtocol containing many #cryptographic vulnerabilities deliberately introduced for educational purposes. - https://l.emre.gs/45PyPhs
11.06.2023 12:36 — 👍 2 🔁 0 💬 0 📌 0
The Dangers of Google’s .zip TLD.
Can you quickly tell which of the URLs below is legitimate and which one is a malicious phish that drops evil.exe? - https://l.emre.gs/3IdPuBm
17.05.2023 10:30 — 👍 2 🔁 0 💬 0 📌 0
Vint Cerf on 3 Mistakes He Made in TCP/IP; The co-creator of the Internet’s protocols admits his crystal ball had a few cracks - https://l.emre.gs/42yMK9K
13.05.2023 20:09 — 👍 1 🔁 0 💬 0 📌 0
18.04.2023 21:43 — 👍 4 🔁 0 💬 0 📌 0
Security Engineering @ Apple, Ex-UK Gov
The intersection of technology (device and national security), and liberal arts (public policy)
DWeb connects people, projects, and protocols essential to building a decentralized web. A web that is private, reliable, secure and open.
DWEB CAMP 2026: July 8-12 near Berlin!
📄 Principles: https://getdweb.net/principles
🌐 Website: https://getdweb.net/
Flibustier du net ̿ ̿̿'̿'\̵͇̿̿\=(•̪●)=/̵͇̿̿/'̿̿ ̿ ̿ ̿
Podcast Hack'n Speak http://anchor.fm/hacknspeak
Github https://github.com/mpgn
Running on vibes. Cyber vibes.
Creators of BloodHound | Experts in Adversary Tradecraft | Leaders in Identity Attack Path Management
I write curl. I don't know anything. I am @bagder@mastodon.social
Weekly email at https://lists.haxx.se/listinfo/daniel
Account for RedTeam Pentesting GmbH
Imprint: https://redteam-pentesting.de/imprint/
Former biologist. Actual PhD in genetics. Security at Nvidia, OS security teaching at https://www.ischool.berkeley.edu.
Blog: https://mjg59.dreamwidth.org
Signal: @mjg.59
He/him
Former Pentester
Engineer at SpecterOps
Author of BloodHound
Advancing human progress together
We’re an independent, nonprofit organization that works side by side with consumers to create a fair and just marketplace.
CEO of Microsoft AI | Author: The Coming Wave | Formerly co-founder at Inflection AI, DeepMind
Global Health. Energy Innovation. Alzheimer’s. https://gatesnot.es/tgn
Chasing digital badness. Senior Researcher at Citizen Lab, but words here are mine.
İfade Özgürlüğü Derneği (İFÖD) | Freedom of Expression Association
ifade.org.tr
Open, transparent AI for real world impact. Built for developers, creators, and teams shaping what’s next.
🔳 founder of @greynoise.io. computers, networks, technology enthusiast. big goober.
Okumak iptilâdır, müptelâlara selam!