From Descriptive to Prescriptive: Uncover the Social Value Alignment of LLM-based Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909041087741952 |
|---|---|
| author | Qu, Jinxian Gu, Qingqing Chen, Teng Ji, Luo |
| author_facet | Qu, Jinxian Gu, Qingqing Chen, Teng Ji, Luo |
| contents | Wide applications of LLM-based agents require strong alignment with human social values. However, current works still exhibit deficiencies in self-cognition and dilemma decision, as well as self-emotions. To remedy this, we propose a novel value-based framework that employs GraphRAG to convert principles into value-based instructions and steer the agent to behave as expected by retrieving the suitable instruction upon a specific conversation context. To evaluate the ratio of expected behaviors, we define the expected behaviors from two famous theories, Maslow's Hierarchy of Needs and Plutchik's Wheel of Emotion. By experimenting with our method on the benchmark of DAILYDILEMMAS, our method exhibits significant performance gains compared to prompt-based baselines, including ECoT, Plan-and-Solve, and Metacognitive prompting. Our method provides a basis for the emergence of self-emotion in AI systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_14034 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | From Descriptive to Prescriptive: Uncover the Social Value Alignment of LLM-based Agents Qu, Jinxian Gu, Qingqing Chen, Teng Ji, Luo Artificial Intelligence Computation and Language Computers and Society Wide applications of LLM-based agents require strong alignment with human social values. However, current works still exhibit deficiencies in self-cognition and dilemma decision, as well as self-emotions. To remedy this, we propose a novel value-based framework that employs GraphRAG to convert principles into value-based instructions and steer the agent to behave as expected by retrieving the suitable instruction upon a specific conversation context. To evaluate the ratio of expected behaviors, we define the expected behaviors from two famous theories, Maslow's Hierarchy of Needs and Plutchik's Wheel of Emotion. By experimenting with our method on the benchmark of DAILYDILEMMAS, our method exhibits significant performance gains compared to prompt-based baselines, including ECoT, Plan-and-Solve, and Metacognitive prompting. Our method provides a basis for the emergence of self-emotion in AI systems. |
| title | From Descriptive to Prescriptive: Uncover the Social Value Alignment of LLM-based Agents |
| topic | Artificial Intelligence Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2605.14034 |