search
HomeTechnology peripheralsAICan fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

Large language models (LLMs) are trained on huge text databases, where they acquire large amounts of real-world knowledge. This knowledge is embedded into their parameters and can then be used when needed. The knowledge of these models is "reified" at the end of training. At the end of pre-training, the model actually stops learning.

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

Align the model or perform instruction tuning to let the model learn how to make full use of this knowledge and how to respond more naturally to the user's questions. But sometimes model knowledge is not enough, and although the model can access external content through RAG, it is considered beneficial to adapt the model to new domains through fine-tuning. This fine-tuning is performed using input created by human annotators or other LLMs, where the model encounters additional real-world knowledge and incorporates it into the parameters.

How does the model integrate this new additional knowledge?

At a mechanistic level, we don’t really know how this interaction occurs of. According to some, exposure to this new knowledge may cause the model to hallucinate. This is because the model is trained to generate facts that are not based on its pre-existing knowledge (or may conflict with the model's prior knowledge). There is also knowledge of what looks the model is likely to encounter (e.g., entities that appear less frequently in the pre-training corpus).

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

#So a recently published study focused on analyzing what happens when a model learns new knowledge through fine-tuning. The authors examine in detail what happens to a fine-tuned model and how it reacts after acquiring new knowledge.

They try to classify the examples at the knowledge level after fine-tuning. The knowledge inherent in a new example may not be completely consistent with the knowledge of the model. Examples can be known or unknown. Even if it is known, it may be highly known, it may be known, or it may be less known knowledge.

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

Then the author used a model (PaLM 2-M) to fine-tune it. Each nudge example is made up of factual knowledge (subjects, relations, objects). This is to allow the model to query this knowledge with specific questions, specific triples (e.g., "Where is Paris?"), and ground truth answers (e.g., "France"). In other words, they provide the model with some new knowledge, and then reconstruct these triples into questions (question-answer pairs) to test its knowledge. They group all these examples into the categories discussed above and then evaluate the answers.

Test results after fine-tuning the model: A high proportion of unknown facts leads to performance degradation (which is not compensated by longer fine-tuning time).

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations


Unknown facts have almost a neutral effect at lower epoch numbers, but at more The number of epochs will hurt performance. So unknown examples appear to be harmful, but their negative impact is mainly reflected in the later stages of training. The graph below shows training accuracy as a function of fine-tuning duration for known and unknown subsets of the dataset example. It can be seen that the model learns unknown examples at a later stage.

Lastly, since Unknown examples are the ones that are likely to introduce new factual knowledge, their significantly slow fitting rate suggests that LLMs struggle to acquire new factual knowledge through fine-tuning , instead they learn to expose their preexisting knowledge using the Known examples. The relationship between examples is quantified and whether it is linear. The results show that there is a strong linear relationship between unknown examples hurting performance and known examples improving performance, almost as strong (the correlation coefficients in this linear regression are very close).

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

这种微调不仅对特定情况下的性能有影响,而且对模型知识有广泛的影响。作者使用分布外(OOD)的测试集表明,未知样本对OOD性能是有害的。根据作者的说法,这与幻觉的发生也有关系:

Overall, our insights transfer across relations. This essentially shows  that fine-tuning on Unknown examples such as “Where is [E1] located?”,  can encourage hallucinations on seemingly unrelated questions, such as  “Who founded [E2]?”.

另外一个有趣的结果是,最好的结果不是用众所周知的例子获得的,而是用可能已知的例子。换句话说,这些例子允许模型更好地利用其先验知识(过于众所周知的事实不会对模型产生有用的影响)。

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

相比之下,未知和不太清楚的事实会损害模型的表现,而这种下降源于幻觉的增加。

This work highlights the risk in using supervised fine-tuning to update  LLMs’ knowledge, as we present empirical evidence that acquiring new  knowledge through finetuning is correlated with hallucinations w.r.t  preexisting knowledge.

根据作者的说法,这种未知的知识可能会损害性能(这使得微调几乎毫无用处)。而用“我不知道”标记这种未知知识可以帮助减少这种伤害。

Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations

Acquiring new knowledge via supervised fine-tuning is correlated with  hallucinations w.r.t. pre-existing knowledge. LLMs struggle to integrate new knowledge through fine-tuning and mostly learn to use their  pre-existing knowledge.

综上所述,如果在微调过程中出现未知知识,则会对模型造成损害。这种性能下降与幻觉的增加有关。相比之下,可能已知的例子反而有有益的影响。这表明该模型难以整合新知识。也就是说在模型所学到的知识和它如何使用新知识之间存在冲突。这可能与对齐和指令调优有关(但是这篇论文没有研究这一点)。

所以如果想要使用具有特定领域知识的模型,论文建议最好使用RAG。并且带有“我不知道”标记的结果可以找到其他策略来克服这些微调的局限性。

这项研究是非常有意思,它表明微调的因素以及如何解决新旧知识之间的冲突仍然不清楚。这就是为什么我们要测试微调前和后结果的原因。

The above is the detailed content of Can fine-tuning really allow LLM to learn new things: introducing new knowledge may make the model produce more hallucinations. For more information, please follow other related articles on the PHP Chinese website!

Statement
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
2023年机器学习的十大概念和技术2023年机器学习的十大概念和技术Apr 04, 2023 pm 12:30 PM

机器学习是一个不断发展的学科,一直在创造新的想法和技术。本文罗列了2023年机器学习的十大概念和技术。 本文罗列了2023年机器学习的十大概念和技术。2023年机器学习的十大概念和技术是一个教计算机从数据中学习的过程,无需明确的编程。机器学习是一个不断发展的学科,一直在创造新的想法和技术。为了保持领先,数据科学家应该关注其中一些网站,以跟上最新的发展。这将有助于了解机器学习中的技术如何在实践中使用,并为自己的业务或工作领域中的可能应用提供想法。2023年机器学习的十大概念和技术:1. 深度神经网

人工智能自动获取知识和技能,实现自我完善的过程是什么人工智能自动获取知识和技能,实现自我完善的过程是什么Aug 24, 2022 am 11:57 AM

实现自我完善的过程是“机器学习”。机器学习是人工智能核心,是使计算机具有智能的根本途径;它使计算机能模拟人的学习行为,自动地通过学习来获取知识和技能,不断改善性能,实现自我完善。机器学习主要研究三方面问题:1、学习机理,人类获取知识、技能和抽象概念的天赋能力;2、学习方法,对生物学习机理进行简化的基础上,用计算的方法进行再现;3、学习系统,能够在一定程度上实现机器学习的系统。

超参数优化比较之网格搜索、随机搜索和贝叶斯优化超参数优化比较之网格搜索、随机搜索和贝叶斯优化Apr 04, 2023 pm 12:05 PM

本文将详细介绍用来提高机器学习效果的最常见的超参数优化方法。 译者 | 朱先忠​审校 | 孙淑娟​简介​通常,在尝试改进机器学习模型时,人们首先想到的解决方案是添加更多的训练数据。额外的数据通常是有帮助(在某些情况下除外)的,但生成高质量的数据可能非常昂贵。通过使用现有数据获得最佳模型性能,超参数优化可以节省我们的时间和资源。​顾名思义,超参数优化是为机器学习模型确定最佳超参数组合以满足优化函数(即,给定研究中的数据集,最大化模型的性能)的过程。换句话说,每个模型都会提供多个有关选项的调整“按钮

得益于OpenAI技术,微软必应的搜索流量超过谷歌得益于OpenAI技术,微软必应的搜索流量超过谷歌Mar 31, 2023 pm 10:38 PM

截至3月20日的数据显示,自微软2月7日推出其人工智能版本以来,必应搜索引擎的页面访问量增加了15.8%,而Alphabet旗下的谷歌搜索引擎则下降了近1%。 3月23日消息,外媒报道称,分析公司Similarweb的数据显示,在整合了OpenAI的技术后,微软旗下的必应在页面访问量方面实现了更多的增长。​​​​截至3月20日的数据显示,自微软2月7日推出其人工智能版本以来,必应搜索引擎的页面访问量增加了15.8%,而Alphabet旗下的谷歌搜索引擎则下降了近1%。这些数据是微软在与谷歌争夺生

荣耀的人工智能助手叫什么名字荣耀的人工智能助手叫什么名字Sep 06, 2022 pm 03:31 PM

荣耀的人工智能助手叫“YOYO”,也即悠悠;YOYO除了能够实现语音操控等基本功能之外,还拥有智慧视觉、智慧识屏、情景智能、智慧搜索等功能,可以在系统设置页面中的智慧助手里进行相关的设置。

人工智能在教育领域的应用主要有哪些人工智能在教育领域的应用主要有哪些Dec 14, 2020 pm 05:08 PM

人工智能在教育领域的应用主要有个性化学习、虚拟导师、教育机器人和场景式教育。人工智能在教育领域的应用目前还处于早期探索阶段,但是潜力却是巨大的。

30行Python代码就可以调用ChatGPT API总结论文的主要内容30行Python代码就可以调用ChatGPT API总结论文的主要内容Apr 04, 2023 pm 12:05 PM

阅读论文可以说是我们的日常工作之一,论文的数量太多,我们如何快速阅读归纳呢?自从ChatGPT出现以后,有很多阅读论文的服务可以使用。其实使用ChatGPT API非常简单,我们只用30行python代码就可以在本地搭建一个自己的应用。 阅读论文可以说是我们的日常工作之一,论文的数量太多,我们如何快速阅读归纳呢?自从ChatGPT出现以后,有很多阅读论文的服务可以使用。其实使用ChatGPT API非常简单,我们只用30行python代码就可以在本地搭建一个自己的应用。使用 Python 和 C

人工智能在生活中的应用有哪些人工智能在生活中的应用有哪些Jul 20, 2022 pm 04:47 PM

人工智能在生活中的应用有:1、虚拟个人助理,使用者可通过声控、文字输入的方式,来完成一些日常生活的小事;2、语音评测,利用云计算技术,将自动口语评测服务放在云端,并开放API接口供客户远程使用;3、无人汽车,主要依靠车内的以计算机系统为主的智能驾驶仪来实现无人驾驶的目标;4、天气预测,通过手机GPRS系统,定位到用户所处的位置,在利用算法,对覆盖全国的雷达图进行数据分析并预测。

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

AI Hentai Generator

AI Hentai Generator

Generate AI Hentai for free.

Hot Article

R.E.P.O. Energy Crystals Explained and What They Do (Yellow Crystal)
2 weeks agoBy尊渡假赌尊渡假赌尊渡假赌
Repo: How To Revive Teammates
1 months agoBy尊渡假赌尊渡假赌尊渡假赌
Hello Kitty Island Adventure: How To Get Giant Seeds
4 weeks agoBy尊渡假赌尊渡假赌尊渡假赌

Hot Tools

MantisBT

MantisBT

Mantis is an easy-to-deploy web-based defect tracking tool designed to aid in product defect tracking. It requires PHP, MySQL and a web server. Check out our demo and hosting services.

VSCode Windows 64-bit Download

VSCode Windows 64-bit Download

A free and powerful IDE editor launched by Microsoft

Dreamweaver Mac version

Dreamweaver Mac version

Visual web development tools

SublimeText3 English version

SublimeText3 English version

Recommended: Win version, supports code prompts!

Notepad++7.3.1

Notepad++7.3.1

Easy-to-use and free code editor