Zhichao Shi

large language model (4) multimodal learning (2) entity alignment (2) knowledge graph (2) harmful content (1) adversarial attack (1) safety evaluation (1) jailbreak attack (1) evaluation metric (1) retrieval-augmented generation (1) knowledge graph alignment (1) multilingual learning (1) llm safety (1) multi-step reasoning (1) natural language to sql (1) adversarial benchmark (1) red team testing (1) execution accuracy (1) toxic content (1) llm detoxification (1)