VLM-Guard: Safeguarding Vision-Language Models via Fulfilling Safety Alignment Gap

Re-assign community