Tags → #large language models

Vulnerabilities
Survey Highlights Security and Privacy Risks of Large Language Models
Nov 19, 2024 A recent survey highlights significant security and privacy threats associated with large language models (LLMs), identifying vulnerabilities such as jailbreaking and data poisoning across various sectors.
Threats
Jailbreaking Techniques and Security Risks in Large Language Models
Nov 17, 2024 The practice of jailbreaking Large Language Models (LLMs) involves manipulating AI systems to produce unintended outputs, raising concerns about potential security vulnerabilities and prompting research into enhancing their security.
Tools
New System LProtector Addresses Cybersecurity Vulnerability Detection Challenges
Nov 15, 2024 Researchers have introduced LProtector, an automated vulnerability detection system designed for C and C++ codebases, which utilizes advanced techniques to improve detection accuracy in complex software systems.
Vulnerabilities
Study Reveals Vulnerabilities in Large Language Models
Nov 15, 2024 A recent study reveals that large language models are vulnerable to jailbreak attacks, prompting researchers to propose a new defense framework called AutoDefense to enhance their security.
Tools
Analysis of Security Scanners for Large Language Models
Nov 14, 2024 The integration of conversational large language models into various applications has prompted a comparative analysis of open-source scanners designed to identify vulnerabilities and enhance security measures.
Tools
Framework Developed for Analyzing Cryptographic Functions in Binaries
Nov 14, 2024 The article discusses the development of a novel framework called FoC, which utilizes large language models to improve the identification and analysis of cryptographic functions in stripped binaries, addressing challenges faced in software security tasks.
Vulnerabilities
New System IRIS Enhances Detection of Software Vulnerabilities
Nov 13, 2024 The development of the IRIS system, which combines large language models with static analysis, aims to enhance the detection of security vulnerabilities in software applications.
Vulnerabilities
Study Reveals Vulnerabilities in Large Language Models
Nov 13, 2024 A recent study has introduced a new framework called DrAttack, which enhances the success rate of jailbreak attacks on Large Language Models by employing a comprehensive approach that includes decomposition, implicit reconstruction, and synonym search.
Vulnerabilities
Concerns Raised Over Safety of Large Language Models
Nov 12, 2024 Recent studies highlight concerns about the safety and ethical implications of large language models (LLMs), revealing vulnerabilities in current unlearning methods that could allow adversaries to access sensitive information.
Industry
Intellectual Property Protection for Deep Learning Models Discussed
Nov 12, 2024 The tech industry is increasingly recognizing the critical importance of Intellectual Property protection for deep learning models and datasets amid rising concerns over unauthorized reproduction and distribution.