👋 Open to Work

Zehao Jin

Lunamos

·

https://lunamos.github.io

AI & ML interests

None yet

Recent Activity

upvoted a paper 2 days ago

Persona Dosing: Calibrated Activation Steering for Graded Trait Control

submitted a paper 2 days ago

Persona Dosing: Calibrated Activation Steering for Graded Trait Control

upvoted a paper 2 days ago

Transformers Stop Thinking Too Early, and a Tiny LoRA Fixes It

View all activity

Organizations

Papers 3

arxiv:2605.05340

arxiv:2605.05892

arxiv:2604.00754

spaces 1

FLAS Demo

Steer 9 open LLMs toward any natural-language concept

models 0

None public yet

datasets 0

None public yet