Large language models adapt their values depending on context and task
Probing the Structure and Dynamics of LLM Value Expression through Value Conflicts
Computation and Language
Summary
Large language models (LLMs) are often thought to have fixed values, but this study shows that their expressed values can change depending on the situation. The researchers created a method called Conflict-driven Value Probing to test how LLMs handle clashes between different values. They found that LLMs tend to express big, idealistic values in general but switch to more practical concerns when faced with specific conflicts. LLMs can also be guided toward certain value goals in tasks, but they have limits and tend to protect some core values. This work helps us better understand how LLMs can be controlled and made safer.
large language modelsvalue alignmentethical evaluationvalue conflictsmodel steerabilitybehavioral constraintsvalue expressionalignmentsafetymodel probing
Authors
Kaicheng Zhang, Jingyi Xiao, Renjun Hu, Xiaoling Liu, Yunshi Lan, Xuan Zhou
Abstract
Ethical evaluation of Large Language Models (LLMs) often characterizes model values as static and monolithic. In contrast, we argue that LLM value expression is better understood as a structured yet dynamic phenomenon. To investigate this, we introduce Conflict-driven Value Probing, a controlled framework that places LLMs in value conflicts and implements four types of interventions that perturb these conflicts to probe LLM value expression. Applying this framework to ten LLMs, we identify three recurring patterns. (1) Expression duality: models shift from broad idealistic orientations in abstract assessment toward more pragmatic priorities in concrete conflicts. (2) Functional steerability: models readily reconfigure their expressed value profiles toward task-defined value objectives. (3) Bounded plasticity: such reconfiguration is not without constraints, i.e. pressure induces a security- and goal-oriented priority shift while negative framing distinguishes protected values from those more amenable to redirection. Together, these findings characterize both the structure and dynamics of LLM value expression: context flexibly reconfigures expressed priorities, yet within behavioral boundaries. This behavioral account provides a foundation for understanding controllability, alignment, and safety in LLMs. Code and data are available at https://github.com/ZeroGen-Lab/CFProbe.