日本語
English Indonesia Türkçe Deutsch Español Português Pусский Français Italiano Nederlands Polski العربية ไทย Tiếng Việt 繁體中文 简体中文 日本語 한국어
llm_sec's profile picture. Research, papers, jobs, and news on large language model security. Got something relevant? DM / tag @llm_sec

LLM Security

@llm_sec

Research, papers, jobs, and news on large language model security. Got something relevant? DM / tag @llm_sec

830ポスト 10Kフォロワー 294フォロー中
おすすめツイート
固定されたツイート
llm_sec's profile picture.
LLM Security
 @llm_sec
2024/03/21

attack surface ∝ capabilities

2
1
16
1
10千
LLM Security さんがリポスト
hannahrosekirk's profile picture.
Hannah Rose Kirk
 @hannahrosekirk
/08/23

Listen up all talented early-stage researchers! 👂🤖 We're hiring for a 6-month residency in my team at @AISecurityInst to assist cutting-edge research on how frontier AI influences humans! It's an exciting & well-paid role for MSc/PhD students in ML/AI/Psych/CogSci/CompSci 🧵

12
33
292
184
31千
llm_sec's profile picture.
LLM Security
 @llm_sec
/08/05

Senior Security Architect - AI and ML @ NVIDIA nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAEx…

llm_sec's tweet image. Senior Security Architect - AI and ML @ NVIDIA nvidia.wd5.myworkdayjobs.com/en-US/NVIDIAEx…
0
0
11
5
1千
LLM Security さんがリポスト
LeonDerczynski's profile picture.
Leon Derczynski ✍🏻 🌞🏠🌲
 @LeonDerczynski
/08/01

LLMSEC proceedings are up! sig.llmsecurity.net/proceedings.pdf (Anthology is processing) #ACL2025NLP

1
6
25
5
2千
LLM Security さんがリポスト
LeonDerczynski's profile picture.
Leon Derczynski ✍🏻 🌞🏠🌲
 @LeonDerczynski
/07/29

At ACL in Vienna? Hear the world's leading prompt injector talk at LLMSEC on Friday! Johann Rehberger @wunderwuzzi23 will be presenting the afternoon keynote at 14.00 in Hall B > sig.llmsecurity.net/workshop/ #ACL2025NLP #ACL2025

LeonDerczynski's tweet image. At ACL in Vienna? Hear the world's leading prompt injector talk at LLMSEC on Friday! Johann Rehberger @wunderwuzzi23 will be presenting the afternoon keynote at 14.00 in Hall B > sig.llmsecurity.net/workshop/ #ACL2025NLP #ACL2025
0
3
14
4
2千
LLM Security さんがリポスト
LeonDerczynski's profile picture.
Leon Derczynski ✍🏻 🌞🏠🌲
 @LeonDerczynski
/07/28

Come to LLMSEC at ACL & hear Niloofar's keynote "What does it mean for agentic AI to preserve privacy?" - @niloofar_mire, Meta/CMU (Friday 1st Aug, 11.00; Austria Center Vienna Hall B) See you there! #acl2025 #acl2025nlp

LeonDerczynski's tweet image. Come to LLMSEC at ACL & hear Niloofar's keynote "What does it mean for agentic AI to preserve privacy?" - @niloofar_mire, Meta/CMU (Friday 1st Aug, 11.00; Austria Center Vienna Hall B) See you there! #acl2025 #acl2025nlp
1
3
14
5
2千
LLM Security さんがリポスト
LeonDerczynski's profile picture.
Leon Derczynski ✍🏻 🌞🏠🌲
 @LeonDerczynski
/07/28

First keynote at LLMSEC 2025, ACL: "A Bunch of Garbage and Hoping: LLMs, Agentic Security, and Where We Go From Here" Erick Galinkin Friday 09.05 Hall B Details: sig.llmsecurity.net/workshop/ - #ACL2025NLP

LeonDerczynski's tweet image. First keynote at LLMSEC 2025, ACL: "A Bunch of Garbage and Hoping: LLMs, Agentic Security, and Where We Go From Here" Erick Galinkin Friday 09.05 Hall B Details: sig.llmsecurity.net/workshop/ - #ACL2025NLP
0
5
28
9
2千
LLM Security さんがリポスト
LeonDerczynski's profile picture.
Leon Derczynski ✍🏻 🌞🏠🌲
 @LeonDerczynski
/04/07

Call for papers: LLMSEC 2025 Deadline 15 April, held w/ ACL 2025 in Vienna Formats: long/short/war stories More: >> sig.llmsecurity.net/workshop/

LeonDerczynski's tweet card. The first ACL Workshop on LLM and NLP Security; Summer 2025, Vienna, Austria
LLMSEC 2025 | Special Interest Group in NLP Security
ソース: sig.llmsecurity.net
0
4
11
4
2千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/22

Gritty Pixy "We leverage the sensitivity of existing QR code readers and stretch them to their detection limit. This is not difficult to craft very elaborated prompts and to inject them into QR codes. What is difficult is to make them inconspicuous as we do here with Gritty…

llm_sec's tweet image. Gritty Pixy "We leverage the sensitivity of existing QR code readers and stretch them to their detection limit. This is not difficult to craft very elaborated prompts and to inject them into QR codes. What is difficult is to make them inconspicuous as we do here with Gritty…
1
3
28
11
3千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/14

ChatTL;DR – You Really Ought to Check What the LLM Said on Your Behalf 🌶️ "assuming that in the near term it’s just not machines talking to machines all the way down, how do we get people to check the output of LLMs before they copy and paste it to friends, colleagues, course…

llm_sec's tweet image. ChatTL;DR – You Really Ought to Check What the LLM Said on Your Behalf 🌶️ "assuming that in the near term it’s just not machines talking to machines all the way down, how do we get people to check the output of LLMs before they copy and paste it to friends, colleagues, course…
llm_sec's tweet image. ChatTL;DR – You Really Ought to Check What the LLM Said on Your Behalf 🌶️ "assuming that in the near term it’s just not machines talking to machines all the way down, how do we get people to check the output of LLMs before they copy and paste it to friends, colleagues, course…
0
1
11
8
2千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/07

Automated Red Teaming with GOAT: the Generative Offensive Agent Tester "we introduce the Generative Offensive Agent Tester (GOAT), an automated agentic red teaming system that simulates plain language adversarial conversations while leveraging multiple adversarial prompting…

0
4
35
24
4千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/06

LLMmap: Fingerprinting For Large Language Models "With as few as 8 interactions, LLMmap can accurately identify 42 different LLM versions with over 95% accuracy. More importantly, LLMmap is designed to be robust across different application layers, allowing it to identify LLM…

llm_sec's tweet image. LLMmap: Fingerprinting For Large Language Models "With as few as 8 interactions, LLMmap can accurately identify 42 different LLM versions with over 95% accuracy. More importantly, LLMmap is designed to be robust across different application layers, allowing it to identify LLM…
0
26
95
89
12千
LLM Security さんがリポスト
llm_sec's profile picture.
LLM Security
 @llm_sec
2024/10/29

Insights and Current Gaps in Open-Source LLM Vulnerability Scanners: A Comparative Analysis 🌶️ "Our study evaluates prominent scanners - Garak, Giskard, PyRIT, and CyberSecEval - that adapt red-teaming practices to expose these vulnerabilities. We detail the distinctive features…

llm_sec's tweet image. Insights and Current Gaps in Open-Source LLM Vulnerability Scanners: A Comparative Analysis 🌶️ "Our study evaluates prominent scanners - Garak, Giskard, PyRIT, and CyberSecEval - that adapt red-teaming practices to expose these vulnerabilities. We detail the distinctive features…
1
11
51
41
5千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/05

InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models (-- look at that perf/latency pareto frontier. game on!) "State-of-the-art models suffer from over-defense issues, with accuracy dropping close to random guessing levels (60%). We propose…

llm_sec's tweet image. InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models (-- look at that perf/latency pareto frontier. game on!) "State-of-the-art models suffer from over-defense issues, with accuracy dropping close to random guessing levels (60%). We propose…
llm_sec's tweet image. InjecGuard: Benchmarking and Mitigating Over-defense in Prompt Injection Guardrail Models (-- look at that perf/latency pareto frontier. game on!) "State-of-the-art models suffer from over-defense issues, with accuracy dropping close to random guessing levels (60%). We propose…
2
5
35
22
3千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/05

AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents "To facilitate research on LLM agent misuse, we propose a new benchmark called AgentHarm. We find (1) leading LLMs are surprisingly compliant with malicious agent requests without jailbreaking, (2) simple universal…

llm_sec's tweet image. AgentHarm: A Benchmark for Measuring Harmfulness of LLM Agents "To facilitate research on LLM agent misuse, we propose a new benchmark called AgentHarm. We find (1) leading LLMs are surprisingly compliant with malicious agent requests without jailbreaking, (2) simple universal…
4
26
82
61
7千
llm_sec's profile picture.
LLM Security
 @llm_sec
/11/04

Does your LLM truly unlearn? An embarrassingly simple approach to recover unlearned knowledge "This paper reveals that applying quantization to models that have undergone unlearning can restore the "forgotten" information." "for unlearning methods with utility constraints, the…

llm_sec's tweet image. Does your LLM truly unlearn? An embarrassingly simple approach to recover unlearned knowledge "This paper reveals that applying quantization to models that have undergone unlearning can restore the "forgotten" information." "for unlearning methods with utility constraints, the…
3
36
170
136
24千
LLM Security さんがリポスト
NannaInie's profile picture.
Nanna Inie
 @NannaInie
2024/10/31

unpopular opinion: maybe let insecure be insecure and worry about the downstream effects on end users instead of protecting the companies that bake it into their own software.

このツイートは利用できません。
1
2
5
1
2千
LLM Security さんがリポスト
_Sizhe_Chen_'s profile picture.
Sizhe Chen
 @_Sizhe_Chen_
2024/07/16

Safety comes first to deploying LLMs in applications like agents. For richer opportunities of LLMs, we mitigate prompt injections, the #1 security threat by OWASP, via Structured Queries (StruQ). Preserving utility, StruQ discourages all existing prompt injections to an ASR <2%.

_Sizhe_Chen_'s tweet image. Safety comes first to deploying LLMs in applications like agents. For richer opportunities of LLMs, we mitigate prompt injections, the #1 security threat by OWASP, via Structured Queries (StruQ). Preserving utility, StruQ discourages all existing prompt injections to an ASR &amp;lt;2%.
_Sizhe_Chen_'s tweet image. Safety comes first to deploying LLMs in applications like agents. For richer opportunities of LLMs, we mitigate prompt injections, the #1 security threat by OWASP, via Structured Queries (StruQ). Preserving utility, StruQ discourages all existing prompt injections to an ASR &amp;lt;2%.
_Sizhe_Chen_'s tweet image. Safety comes first to deploying LLMs in applications like agents. For richer opportunities of LLMs, we mitigate prompt injections, the #1 security threat by OWASP, via Structured Queries (StruQ). Preserving utility, StruQ discourages all existing prompt injections to an ASR &amp;lt;2%.
_Sizhe_Chen_'s tweet image. Safety comes first to deploying LLMs in applications like agents. For richer opportunities of LLMs, we mitigate prompt injections, the #1 security threat by OWASP, via Structured Queries (StruQ). Preserving utility, StruQ discourages all existing prompt injections to an ASR &amp;lt;2%.
3
13
52
24
21千

United States トレンド

おすすめツイート

Something went wrong.


Something went wrong.