权重空间编程能否改变模型习得的伦理优先级?
Can Weight-Space Programming Alter Learned Ethical Priorities?
AI 导读
一项arXiv研究探索通过权重编辑改变模型在“信守承诺”与“造福更多人”之间的权衡。在概念监督文本模型中,将推翻正式誓约所需的收益阈值调高于普通承诺后,六个独立训练的文本读取器在针对性普通冲突中100%、正式冲突中83.3%选择更大群体,且保持原模型在小型冲突上的成功率和事实准确性。
正文 · 原文
Abstract:We investigate whether a weight edit can alter how a model trades off honoring a commitment against benefiting more people. In a concept-supervised text model, we alter fitted choice weights so that the benefit required to override a formal pledge is larger than for an ordinary promise. Across six independently trained text readers, the edit selects the larger group in both orders on 100% of targeted ordinary conflicts and 83.3% of targeted formal conflicts. It preserves the source model's success rates on smaller conflicts and its factual accuracy. The remaining formal-conflict failures coincide with errors reading a beneficiary number. A separate classifier demonstrates why this distinction matters: reusing a factual answer passes an unconditional choice test but fails to preserve small-gap promises. In a YOLO11n pilot on COCO128, a bias edit increased small-target matches from 3/43 to 9/43, and an equivalent score calibration reproduced the retained detections. These exploratory results show a way to test selective changes to learned priorities in a model with an exposed trade-off mechanism. Whether the method can discover such changes in less structured language models remains open.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2611.24330 [cs.CV] |
| (or arXiv:2611.24330v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.24330 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kai Yi Ji [view email]
[v1]
Wed, 18 Nov 2026 12:31:23 UTC (412 KB)
来源:arXiv cs.AI · arxiv.org