arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2305.00593cs.LGcs.CL
收藏

可靠的无梯度且无似然提示调优

Reliable Gradient-free and Likelihood-free Prompt Tuning

Massachusetts Institute of Technology(麻省理工学院);MIT-IBM Watson AI Lab(麻省理工学院-IBM沃森人工智能实验室);IBM Research(IBM研究院);University of Florida(佛罗里达大学)

Maohao Shen, Soumya Ghosh, Prasanna Sattigeri, Subhro Das, Yuheng Bu, Gregory Wornell

AI总结:

针对黑盒API形式的大型预训练语言模型难以微调的问题,提出无梯度、无似然的软提示调优方法,通过学习提示分布量化不确定性,性能媲美甚至优于全访问的梯度方法。

摘要:

由于隐私或商业限制,大型预训练语言模型(PLMs)通常以黑盒API的形式提供。对这类模型进行下游任务微调颇具挑战性,因为研究者既无法访问模型的内部表示,也无法通过模型传播梯度。本文针对这些挑战,开发了仅通过API访问即可适配PLMs的技术。基于近期软提示调优的相关研究,我们提出了无需梯度计算即可调优软提示的方法。此外,我们还对方法进行了扩展,这类扩展方案除无需梯度外,除输入嵌入外也不需要访问PLM的任何内部表示。不仅如此,我们的方法并非学习单个提示,而是学习提示上的分布,从而能够量化预测不确定性。我们的工作是首个在仅能通过API访问PLM的场景下,研究提示中不确定性的工作。最后,通过大量实验,我们对所提出的方法进行了细致验证,结果表明这些方法的性能与可完全访问PLM的基于梯度的方法相当,有时甚至更优。

英文摘要:

Due to privacy or commercial constraints, large pre-trained language models (PLMs) are often offered as black-box APIs. Fine-tuning such models to downstream tasks is challenging because one can neither access the model's internal representations nor propagate gradients through it. This paper addresses these challenges by developing techniques for adapting PLMs with only API access. Building on recent work on soft prompt tuning, we develop methods to tune the soft prompts without requiring gradient computation. Further, we develop extensions that in addition to not requiring gradients also do not need to access any internal representation of the PLM beyond the input embeddings. Moreover, instead of learning a single prompt, our methods learn a distribution over prompts allowing us to quantify predictive uncertainty. Ours is the first work to consider uncertainty in prompts when only having API access to the PLM. Finally, through extensive experiments, we carefully vet the proposed methods and find them competitive with (and sometimes even improving on) gradient-based approaches with full access to the PLM.

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。

↑