most citedLearning to Reason for Hallucination Span Detection

1 citations · 1 across the 3 of their papers we have counts for

collaborators
Showing cs.LGShow all

1 paper · 1 filter

cs.LG2025

Disentangled Safety Adapters Enable Efficient Guardrails and Flexible Inference-Time Alignment

Kundan Krishna, Joseph Y Cheng, Charles Maalouf +1

Existing paradigms for ensuring AI safety, such as guardrail models and alignment training, often compromise either inference efficiency or development flexibility. We introduce Di…