Following the Whispers of Values: Unraveling Neural Mechanisms Behind Value-Oriented Behaviors in LLMs
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913800816427008 |
|---|---|
| author | Hu, Ling Xu, Yuemei Gu, Xiaoyang Han, Letao |
| author_facet | Hu, Ling Xu, Yuemei Gu, Xiaoyang Han, Letao |
| contents | Despite the impressive performance of large language models (LLMs), they can present unintended biases and harmful behaviors driven by encoded values, emphasizing the urgent need to understand the value mechanisms behind them. However, current research primarily evaluates these values through external responses with a focus on AI safety, lacking interpretability and failing to assess social values in real-world contexts. In this paper, we propose a novel framework called ValueExploration, which aims to explore the behavior-driven mechanisms of National Social Values within LLMs at the neuron level. As a case study, we focus on Chinese Social Values and first construct C-voice, a large-scale bilingual benchmark for identifying and evaluating Chinese Social Values in LLMs. By leveraging C-voice, we then identify and locate the neurons responsible for encoding these values according to activation difference. Finally, by deactivating these neurons, we analyze shifts in model behavior, uncovering the internal mechanism by which values influence LLM decision-making. Extensive experiments on four representative LLMs validate the efficacy of our framework. The benchmark and code will be available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_04994 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Following the Whispers of Values: Unraveling Neural Mechanisms Behind Value-Oriented Behaviors in LLMs Hu, Ling Xu, Yuemei Gu, Xiaoyang Han, Letao Computation and Language Artificial Intelligence Despite the impressive performance of large language models (LLMs), they can present unintended biases and harmful behaviors driven by encoded values, emphasizing the urgent need to understand the value mechanisms behind them. However, current research primarily evaluates these values through external responses with a focus on AI safety, lacking interpretability and failing to assess social values in real-world contexts. In this paper, we propose a novel framework called ValueExploration, which aims to explore the behavior-driven mechanisms of National Social Values within LLMs at the neuron level. As a case study, we focus on Chinese Social Values and first construct C-voice, a large-scale bilingual benchmark for identifying and evaluating Chinese Social Values in LLMs. By leveraging C-voice, we then identify and locate the neurons responsible for encoding these values according to activation difference. Finally, by deactivating these neurons, we analyze shifts in model behavior, uncovering the internal mechanism by which values influence LLM decision-making. Extensive experiments on four representative LLMs validate the efficacy of our framework. The benchmark and code will be available. |
| title | Following the Whispers of Values: Unraveling Neural Mechanisms Behind Value-Oriented Behaviors in LLMs |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2504.04994 |