有用性
计算机科学
背景(考古学)
一致性(知识库)
机器学习
人工智能
匹配(统计)
依赖关系(UML)
水准点(测量)
概率逻辑
选择(遗传算法)
人机交互
构造(python库)
成对比较
控制(管理)
语言模型
偏爱
功能(生物学)
国家(计算机科学)
功能可见性
心理学
忠诚
自治
作者
Shalima Binta Manir,Tim Oates
标识
DOI:10.48550/arxiv.2604.01576
摘要
Large language models deployed in supportive or advisory roles must balance helpfulness with preservation of user autonomy, yet standard alignment methods primarily optimize for helpfulness and harmlessness without explicitly modeling relational risks such as dependency reinforcement, overprotection, or coercive guidance. We introduce Care-Conditioned Neuromodulation (CCN), a state-dependent control framework in which a learned scalar signal derived from structured user state and dialogue context conditions response generation and candidate selection. We formalize this setting as an autonomy-preserving alignment problem and define a utility function that rewards autonomy support and helpfulness while penalizing dependency and coercion. We also construct a benchmark of relational failure modes in multi-turn dialogue, including reassurance dependence, manipulative care, overprotection, and boundary inconsistency. On this benchmark, care-conditioned candidate generation combined with utility-based reranking improves autonomy-preserving utility by +0.25 over supervised fine-tuning and +0.07 over preference optimization baselines while maintaining comparable supportiveness. Pilot human evaluation and zero-shot transfer to real emotional-support conversations show directional agreement with automated metrics. These results suggest that state-dependent control combined with utility-based selection is a practical approach to multi-objective alignment in autonomy-sensitive dialogue.
科研通智能强力驱动
Strongly Powered by AbleSci AI