基于变压器的语言模型能够生成流利的文本，并在各种自然语言生成任务中有效地适应。但是，已证明在大型未标记的网络文本语料库中鉴定的语言模型已被证明会遭受堕落的有毒内容和社会偏见行为的损害，从而阻碍了他们的安全部署。提出了各种排毒方法来减轻语言模型的毒性；但是，这些方法是在包含与性别，种族或宗教相关的特定社会身份的提示条件下进行排毒语言模型的。在这项研究中，我们提出了增强氧化。一种基于强化学习的方法，用于降低语言模型中的毒性。我们应对语言模型中的安全性挑战，并提出了一种新的奖励模型，该模型能够检测有毒内容并减轻对毒性预测中社会身份的意外偏见。该实验表明，用于语言模型排毒的增强方法化方法优于自动评估指标中现有的排毒方法，这表明我们在语言模型排毒中的方法能力和对生成内容中社会认同的意外偏见的能力较小。

Steering language generation towards objectives or away from undesired content has been a long-standing goal in utilizing language models (LM). Recent work has demonstrated reinforcement learning and weighted decoding as effective approaches to achieve a higher level of language control and quality with pros and cons. In this work, we propose a novel critic decoding method for controlled language generation (CriticControl) that combines the strengths of reinforcement learning and weighted decoding. Specifically, we adopt the actor-critic framework to train an LM-steering critic from non-differentiable reward models. And similar to weighted decoding, our method freezes the language model and manipulates the output token distribution using called critic, improving training efficiency and stability. Evaluation of our method on three controlled generation tasks, namely topic control, sentiment control, and detoxification, shows that our approach generates more coherent and well-controlled texts than previous methods. In addition, CriticControl demonstrates superior generalization ability in zero-shot settings. Human evaluation studies also corroborate our findings.

Large pretrained language models generate fluent text but are notoriously hard to controllably sample from. In this work, we study constrained sampling from such language models: generating text that satisfies user-defined constraints, while maintaining fluency and the model's performance in a downstream task. We propose MuCoLa -- a sampling procedure that combines the log-likelihood of the language model with arbitrary (differentiable) constraints in a single energy function, and then generates samples in a non-autoregressive manner. Specifically, it initializes the entire output sequence with noise and follows a Markov chain defined by Langevin Dynamics using the gradients of the energy function. We evaluate MuCoLa on text generation with soft and hard constraints as well as their combinations obtaining significant improvements over competitive baselines for toxicity avoidance, sentiment control, and keyword-guided generation.

Controllable Text Generation (CTG) is emerging area in the field of natural language generation (NLG). It is regarded as crucial for the development of advanced text generation technologies that are more natural and better meet the specific constraints in practical applications. In recent years, methods using large-scale pre-trained language models (PLMs), in particular the widely used transformer-based PLMs, have become a new paradigm of NLG, allowing generation of more diverse and fluent text. However, due to the lower level of interpretability of deep neural networks, the controllability of these methods need to be guaranteed. To this end, controllable text generation using transformer-based PLMs has become a rapidly growing yet challenging new research hotspot. A diverse range of approaches have emerged in the recent 3-4 years, targeting different CTG tasks which may require different types of controlled constraints. In this paper, we present a systematic critical review on the common tasks, main approaches and evaluation methods in this area. Finally, we discuss the challenges that the field is facing, and put forward various promising future directions. To the best of our knowledge, this is the first survey paper to summarize CTG techniques from the perspective of PLMs. We hope it can help researchers in related fields to quickly track the academic frontier, providing them with a landscape of the area and a roadmap for future research.

深度神经语言模型的最新进展与大规模数据集的能力相结合，加速了自然语言生成系统的发展，这些系统在多种任务和应用程序上下文中产生流利和连贯的文本（在各种成功程度上）。但是，为所需的用户控制这些模型的输出仍然是一个开放的挑战。这不仅对于自定义生成语言的内容和样式至关重要，而且对于他们在现实世界中的安全可靠部署至关重要。我们提出了一项关于受约束神经语言生成的新兴主题的广泛调查，在该主题中，我们通过区分条件和约束（后者是在输出文本上而不是输入的可检验条件），正式定义和分类自然语言生成问题，目前是可检验的）约束文本生成任务，并查看受限文本生成的现有方法和评估指标。我们的目的是强调这个新兴领域的最新进展和趋势，以告知最有希望的方向和局限性，以推动受约束神经语言生成研究的最新作品。

大型语言模型会产生类似人类的文本，这些文本推动了越来越多的应用。但是，最近的文献以及越来越多的现实世界观察表明，这些模型可以产生有毒，有偏见，不真实或其他有害的语言。尽管正在进行评估语言模型危害的工作，但要远见卓识转换出可能出现的危害可能会引起严格的基准。为了促进这种翻译，我们概述了六种表征有害文本的方式，这些方法在设计新基准时值得明确考虑。然后，我们将这些特征用作镜头来识别现有基准中的趋势和差距。最后，我们将它们应用于视角API的案例研究，这是一种毒性分类器，被广泛用于HARS基准。我们的特征提供了一块桥梁，可以在远见和有效评估之间转化。

Transformer, originally devised for natural language processing, has also attested significant success in computer vision. Thanks to its super expressive power, researchers are investigating ways to deploy transformers to reinforcement learning (RL) and the transformer-based models have manifested their potential in representative RL benchmarks. In this paper, we collect and dissect recent advances on transforming RL by transformer (transformer-based RL or TRL), in order to explore its development trajectory and future trend. We group existing developments in two categories: architecture enhancement and trajectory optimization, and examine the main applications of TRL in robotic manipulation, text-based games, navigation and autonomous driving. For architecture enhancement, these methods consider how to apply the powerful transformer structure to RL problems under the traditional RL framework, which model agents and environments much more precisely than deep RL methods, but they are still limited by the inherent defects of traditional RL algorithms, such as bootstrapping and "deadly triad". For trajectory optimization, these methods treat RL problems as sequence modeling and train a joint state-action model over entire trajectories under the behavior cloning framework, which are able to extract policies from static datasets and fully use the long-sequence modeling capability of the transformer. Given these advancements, extensions and challenges in TRL are reviewed and proposals about future direction are discussed. We hope that this survey can provide a detailed introduction to TRL and motivate future research in this rapidly developing field.

当前的语言模型达到了较低的困惑，但其产生的几代人仍然遭受有毒的反应，重复性和矛盾。标准语言建模设置无法解决这些问题。在本文中，我们介绍了一个新的体系结构{\ sc导演}，由一个统一的生成器分类器组成，具有语言建模和每个输出令牌的分类头。培训是使用标准语言建模数据共同进行的，并以所需和不良序列标记的数据。与标准语言模型相比，该模型在多种设置中的实验表明，该模型具有竞争性的培训和解码速度，同时产生了较高的结果，从而减轻了已知的问题，同时保持发电质量。就准确性和效率而言，它还优于现有的模型指导方法。

We present Second Thought, a new learning paradigm that enables language models (LMs) to re-align with human values. By modeling the chain-of-edits between value-unaligned and value-aligned text, with LM fine-tuning and additional refinement through reinforcement learning, Second Thought not only achieves superior performance in three value alignment benchmark datasets but also shows strong human-value transfer learning ability in few-shot scenarios. The generated editing steps also offer better interpretability and ease for interactive error correction. Extensive human evaluations further confirm its effectiveness.

预处理的基于变压器的语言模型（LMS）显示出显着的自然语言生成能力。凭借其巨大的潜力，控制这种LM的文本生成引起了人们的关注。尽管有一些研究试图控制生成的文本的高级属性（例如情感和主题），但仍然缺乏对其在单词和短语级别上的内容的更精确的控制。在这里，我们建议内容调节器（COCON）以细粒度的水平控制LM的输出文本。在我们的自我监督方法中，Cocon Block学会了通过调节从LM中扣留的内容输入来帮助LM完成部分观察到的文本序列。通过实验，我们表明Cocon可以自然地将目标内容纳入生成的文本中，并以零拍的方式控制高级文本属性。

This paper presents a new UNIfied pre-trained Language Model (UNILM) that can be fine-tuned for both natural language understanding and generation tasks. The model is pre-trained using three types of language modeling tasks: unidirectional, bidirectional, and sequence-to-sequence prediction. The unified modeling is achieved by employing a shared Transformer network and utilizing specific self-attention masks to control what context the prediction conditions on. UNILM compares favorably with BERT on the GLUE benchmark, and the SQuAD 2.0 and CoQA question answering tasks. Moreover, UNILM achieves new state-ofthe-art results on five natural language generation datasets, including improving the CNN/DailyMail abstractive summarization ROUGE-L to 40.51 (2.04 absolute improvement), the Gigaword abstractive summarization ROUGE-L to 35.75 (0.86 absolute improvement), the CoQA generative question answering F1 score to 82.5 (37.1 absolute improvement), the SQuAD question generation BLEU-4 to 22.12 (3.75 absolute improvement), and the DSTC7 document-grounded dialog response generation NIST-4 to 2.67 (human performance is 2.65). The code and pre-trained models are available at https://github.com/microsoft/unilm. * Equal contribution. † Contact person.

Well-designed prompts can guide text-to-image models to generate amazing images. However, the performant prompts are often model-specific and misaligned with user input. Instead of laborious human engineering, we propose prompt adaptation, a general framework that automatically adapts original user input to model-preferred prompts. Specifically, we first perform supervised fine-tuning with a pretrained language model on a small collection of manually engineered prompts. Then we use reinforcement learning to explore better prompts. We define a reward function that encourages the policy to generate more aesthetically pleasing images while preserving the original user intentions. Experimental results on Stable Diffusion show that our method outperforms manual prompt engineering in terms of both automatic metrics and human preference ratings. Moreover, reinforcement learning further boosts performance, especially on out-of-domain prompts. The pretrained checkpoints are available at https://aka.ms/promptist. The demo can be found at https://aka.ms/promptist-demo.

许多文献表明，基于及时的学习是使用大型预训练的语言模型的有效方法。最近的作品还展示了通过插入适当的提示来指导聊天机器人输出的可能性。基于梯度的方法通常用于扰动提示。但是，某些语言模型甚至无法为公众提供。在这项工作中，我们首先探讨了提示和加强学习（RL）与转向模型的生成的组合，而无需访问任何模型的参数。其次，为了减少培训工作并增强对看不见的任务的普遍性，我们应用多任务学习以使模型学会更好地对新任务进行推广。实验结果表明，我们提出的方法可以成功控制几个最新的（SOTA）对话模型，而无需访问其参数。此外，该模型证明了与基线模型更少的步骤快速适应看不见的任务的强大能力。

聊天机器人用于许多应用程序中，例如自动化代理，智能家庭助理，在线游戏中的互动角色等。因此，确保他们不会以不希望的方式行事，对用户提供令人反感或有毒的反应。这并不是一项琐碎的任务，因为最先进的聊天机器人模型是在从互联网公开收集的大型公共数据集上培训的。本文提出了对聊天机器人中毒性的首次大规模测量。我们表明，公开可用的聊天机器人很容易在喂养有毒的查询时提供有毒的反应。更令人担忧的是，一些无毒的查询也会触发有毒反应。然后，我们着手设计和实验攻击，即毒性，该攻击依赖于微调的GPT-2来产生无毒的查询，使聊天机器人以有毒的方式做出反应。我们广泛的实验评估表明，我们的攻击对公共聊天机器人模型有效，并且优于先前工作提出的手动制作的恶意查询。我们还评估了针对毒性的三种防御机制，表明它们要么以影响聊天机器人的效用而降低攻击性能，要么仅有效地减轻了一部分攻击。这强调了对计算机安全和在线安全社区进行更多研究的需求，以确保聊天机器人模型不会伤害其用户。总体而言，我们有信心有毒可以用作审计工具，我们的工作将为设计更有效的聊天机器人安全防御措施铺平道路。

Text detoxification has the potential to mitigate the harms of toxicity by rephrasing text to remove offensive meaning, but subtle toxicity remains challenging to tackle. We introduce MaRCo, a detoxification algorithm that combines controllable generation and text rewriting methods using a Product of Experts with autoencoder language models (LMs). MaRCo uses likelihoods under a non-toxic LM (expert) and a toxic LM (anti-expert) to find candidate words to mask and potentially replace. We evaluate our method on several subtle toxicity and microaggressions datasets, and show that it not only outperforms baselines on automatic metrics, but MaRCo's rewrites are preferred 2.1 $\times$ more in human evaluation. Its applicability to instances of subtle toxicity is especially promising, demonstrating a path forward for addressing increasingly elusive online hate.

本次调查绘制了用于分析社交媒体数据的生成方法的研究状态的广泛的全景照片（Sota）。它填补了空白，因为现有的调查文章在其范围内或被约会。我们包括两个重要方面，目前正在挖掘和建模社交媒体的重要性：动态和网络。社会动态对于了解影响影响或疾病的传播，友谊的形成，友谊的形成等，另一方面，可以捕获各种复杂关系，提供额外的洞察力和识别否则将不会被注意的重要模式。

It has become common to publish large (billion parameter) language models that have been trained on private datasets. This paper demonstrates that in such settings, an adversary can perform a training data extraction attack to recover individual training examples by querying the language model. We demonstrate our attack on GPT-2, a language model trained on scrapes of the public Internet, and are able to extract hundreds of verbatim text sequences from the model's training data. These extracted examples include (public) personally identifiable information (names, phone numbers, and email addresses), IRC conversations, code, and 128-bit UUIDs. Our attack is possible even though each of the above sequences are included in just one document in the training data.We comprehensively evaluate our extraction attack to understand the factors that contribute to its success. Worryingly, we find that larger models are more vulnerable than smaller models. We conclude by drawing lessons and discussing possible safeguards for training large language models.

程序合成或代码生成旨在生成满足问题规范的程序。使用大规模预处理的语言模型（LMS）的最新方法显示出令人鼓舞的结果，但它们有一些关键的局限性。特别是，他们经常遵循标准监督的微调程序，仅从对自然语言问题描述和基础真相计划对培训代码生成模型。这种范式在很大程度上忽略了问题规范中的一些重要但潜在的信号，例如单位测试，因此在求解复杂的看不见的编码任务时通常会导致性能差。为了解决这些局限性，我们提出了“ Coderl”，这是通过验证的LMS和深入强化学习（RL）实现程序合成任务的新框架。具体而言，在培训期间，我们将代码生成的LM视为参与者网络，并引入批评网络，该网络经过培训，以预测生成的程序的功能正确性，并为演员提供密集的反馈信号。在推理期间，我们引入了一种新一代程序，具有关键的抽样策略，该过程允许模型根据示例单位测试和评论家分数的反馈自动重新生成程序。对于模型骨架，我们扩展了Codet5的编码器架构，具有增强的学习目标，更大的模型大小和更好的预处理数据。我们的方法不仅在具有挑战性的应用程序基准上实现了新的SOTA结果，而且还显示出强大的零弹性传输能力，并在简单的MBPP基准上具有新的SOTA结果。

Task agnostic generative pretraining (GPT) has recently proved promising for zero- and few-shot learning, gradually diverting attention from the expensive supervised learning paradigm. Although the community is accumulating knowledge as to capabilities of English-language autoregressive models such as GPT-3 adopting this generative approach, scholarship about these models remains acutely Anglocentric. Consequently, the community currently has serious gaps in its understanding of this class of models, their potential, and their societal impacts in diverse settings, linguistic traditions, and cultures. To alleviate this issue for Arabic, a collection of diverse languages and language varieties with more than $400$ million population, we introduce JASMINE, a suite of powerful Arabic autoregressive Transformer language models ranging in size between 300 million-13 billion parameters. We pretrain our new models with large amounts of diverse data (400GB of text) from different Arabic varieties and domains. We evaluate JASMINE extensively in both intrinsic and extrinsic settings, using a comprehensive benchmark for zero- and few-shot learning across a wide range of NLP tasks. We also carefully develop and release a novel benchmark for both automated and human evaluation of Arabic autoregressive models focused at investigating potential social biases, harms, and toxicity in these models. We aim to responsibly release our models with interested researchers, along with code for experimenting with them

We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations from Transformers. Unlike recent language representation models (Peters et al., 2018a;Radford et al., 2018), BERT is designed to pretrain deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result, the pre-trained BERT model can be finetuned with just one additional output layer to create state-of-the-art models for a wide range of tasks, such as question answering and language inference, without substantial taskspecific architecture modifications.BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural language processing tasks, including pushing the GLUE score to 80.5% (7.7% point absolute improvement), MultiNLI accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).