arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2402.14433cs.CLcs.AI
收藏

语言模型穿越潜在空间的指南

A Language Model's Guide Through Latent Space

ETH Zurich(苏黎世联邦理工学院)

Dimitri von Rütte, Sotiris Anagnostidis, Gregor Bachmann, Thomas Hofmann

AI总结:

本文将概念引导从真实性扩展到得体性、幽默等更复杂概念,提出兼顾激发成功与流畅性的新度量,发现检测最优的探针未必是最佳引导器,并构建了引导研究实验平台。

摘要:

概念引导已成为一种廉价且简单的方法,通过探测语言模型隐藏表征中的概念向量,并在推理时利用这些向量扰动激活,从而控制模型行为。以往研究主要关注真实性,本文则将该框架扩展到更丰富的概念集合,如得体性、幽默、创造性和质量,并探索当前的检测与引导策略在这些具有挑战性的场景中的有效程度。为便于评估,我们提出了一种新的概念引导度量指标,该指标同时考虑概念激发是否成功以及被引导模型流畅性可能出现的退化。大量实验表明,虽然真实性等概念在当前技术下较易实现引导,但得体性或幽默等新概念要么难以激发,要么需要大量调参才能生效,甚至会出现概念混淆。此外,我们发现检测准确率最优的探针并不一定能成为最优的引导器,这与之前关于真实性的观察结果相矛盾。我们的工作表明有必要更深入地研究可检测性、可引导性与概念本质之间的相互作用,并希望我们为引导研究构建的丰富实验测试平台能够激发更强的后续方法。

英文摘要:

Concept guidance has emerged as a cheap and simple way to control the behavior of language models by probing their hidden representations for concept vectors and using them to perturb activations at inference time. While the focus of previous work has largely been on truthfulness, in this paper we extend this framework to a richer set of concepts such as appropriateness, humor, creativity and quality, and explore to what degree current detection and guidance strategies work in these challenging settings. To facilitate evaluation, we develop a novel metric for concept guidance that takes into account both the success of concept elicitation as well as the potential degradation in fluency of the guided model. Our extensive experiments reveal that while some concepts such as truthfulness more easily allow for guidance with current techniques, novel concepts such as appropriateness or humor either remain difficult to elicit, need extensive tuning to work, or even experience confusion. Moreover, we find that probes with optimal detection accuracies do not necessarily make for the optimal guides, contradicting previous observations for truthfulness. Our work warrants a deeper investigation into the interplay between detectability, guidability, and the nature of the concept, and we hope that our rich experimental test-bed for guidance research inspires stronger follow-up approaches.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑