作者位次 #8/13ACL 2026Workshop· ACL System Demonstrations
AI Steerability 360: A Toolkit for Steering Large Language Models
Erik Miehling · Karthikeyan Natesan Ramamurthy · Praveen Venkateswaran · Ching-Yun Ko · Pierre Dognin · Moninder Singh · Tejaswini Pedapati · Avinash Balakrishnan · Matthew Riemer · Dennis Wei · Inge Vejsbjerg · Elizabeth M. Daly · Kush R. Varshney
摘要
The AI Steerability 360 toolkit is an extensible, open-source Python library for steering LLMs. Steering abstractions are designed around four model control surfaces: input (modification of the prompt), structural (modification of the model’s weights or architecture), state (modification of the m...