These 4 critical AI vulnerabilities are being exploited faster than defenders can respond

From prompt injection to deepfake fraud, security researchers say several flaws have no known fix. Here's what to know about them.
GitHub

SQL-R1: Training Natural Language to SQL Reasoning Model By Reinforcement Learning

Natural Language to SQL (NL2SQL) enables intuitive interactions with databases by transforming natural language queries into structured SQL statements. Despite recent advancements in enhancing ...
IEEE

Task Knowledge Injection: Training-Free Adaptation of Multimodal Large Language Models for Remote Sensing Image Understanding

Abstract: Parameter fine-tuning is the mainstream approach for adapting multimodal large language models (MLLMs) to downstream remote sensing tasks. However, such a method risks degrading pretrained ...
IEEE

SQL Injection in LLM-Generated Queries: Systematic Analysis of Detection Gaps and Security Risks

Abstract: Large language models (LLMs) are being woven into software systems at a remarkable pace. When these systems include a back-end database, LLM integration opens new attack surfaces for SQL ...