#safety post-training
Safety Post-Training: 1 AI articles covering safety post-training news, analysis, and research
Articles
DeflectBench: A Benchmark for Evaluating Rhetorical Fallacy Generation in LLMsNEW⭐9
Evaluate LLM rhetorical fallacy generation with DeflectBench: prompt framing drives refusal more than claim content, revealing critical safety vulnerabilities.