arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2305.01628cs.CLcs.LG
收藏

“坏建议”的益处:跨模型层的自对比解码

The Benefits of Bad Advice: Autocontrastive Decoding across Model Layers

IBM Research(IBM研究院)

Ariel Gera, Roni Friedman, Ofir Arviv, Chulaka Gunasekara, Benjamin Sznajder, Noam Slonim, Eyal Shnarch

AI总结:

针对语言模型通常仅用最终层表征、中间层信息被忽视的问题,提出利用高低层预测对比的自对比解码方法,缓解开放式生成退化、提升文本质量,还能更高效提取模型参数中的知识。

摘要:

将语言模型应用于自然语言处理任务通常依赖模型最终层的表征,因为中间隐藏层的表征被认为信息量较低。在本研究中,我们认为由于模型各层的性能是逐步提升的,推理过程中可以从高层与低层的对比中挖掘额外信息。具体而言,在生成模型的多个高概率下一词元预测结果中做选择时,低层的预测可用于凸显哪些候选词元最应该被规避。我们提出了一种利用层间对比提升文本生成输出效果的新方法,并证明该方法缓解了模型在开放式生成中的退化行为,显著提升了生成文本的质量。此外,我们的结果表明,在推理时对比模型各层能为通用语言模型能力的特定方面带来实质性增益,可在推理时从给定的模型参数集中更有效地提取知识。

英文摘要:

Applying language models to natural language processing tasks typically relies on the representations in the final model layer, as intermediate hidden layer representations are presumed to be less informative. In this work, we argue that due to the gradual improvement across model layers, additional information can be gleaned from the contrast between higher and lower layers during inference. Specifically, in choosing between the probable next token predictions of a generative model, the predictions of lower layers can be used to highlight which candidates are best avoided. We propose a novel approach that utilizes the contrast between layers to improve text generation outputs, and show that it mitigates degenerative behaviors of the model in open-ended generation, significantly improving the quality of generated texts. Furthermore, our results indicate that contrasting between model layers at inference time can yield substantial benefits to certain aspects of general language model capabilities, more effectively extracting knowledge during inference from a given set of model parameters.

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。

↑