Efficient LLMs

klieberman

Activity Feed Request to join this org

AI & ML interests

None defined yet.

Recent Activity

jyhong836 authored a paper about 6 hours ago

Safe and Robust Watermark Injection with a Single OoD Image

jyhong836 authored a paper about 6 hours ago

LoX: Low-Rank Extrapolation Robustifies LLM Safety Against Fine-tuning

jyhong836 authored a paper 5 months ago

MedHallu: A Comprehensive Benchmark for Detecting Medical Hallucinations in Large Language Models

View all activity

models 0

None public yet

datasets 0

None public yet