Tag
New research reveals multimodal LLMs are vulnerable to hidden instructions embedded in images. Mind maps, steganography, and physical signage all bypass text-based safety filters.
Run image analysis, document OCR, and visual reasoning locally. Qwen3-VL, Gemma 3, Phi-4 Vision, and more tested from 8GB to 32GB VRAM with real benchmarks.
Alibaba's new 0.8B to 9B parameter models deliver GPT-class multimodal performance on consumer hardware, with the 9B variant outperforming models 13 times its size
China's DeepSeek is releasing V4 - a trillion-parameter multimodal model optimized for domestic chips - while blocking US chipmakers and facing distillation accusations from OpenAI and Anthropic.
A study of 82,000 harm ratings across eight model releases finds 'alignment drift': GPT-5 and Claude 4.5 are more vulnerable to adversarial attacks than their predecessors.
