As language models (LMs) scale, they develop many novel behaviors, good and bad, exacerbating the need to evaluate how they behave. Prior work creates evaluations with crowdwork (which is time-consuming and expensive) or existing data sources (which are not always available). Here, we automatically generate evaluations with LMs. We explore approaches with varying amounts of human effort, from instructing LMs to write yes/no questions to making complex Winogender schemas with multiple stages of LM-based generation and filtering. Crowdworkers rate the examples as highly relevant and agree with 90-100% of labels, sometimes more so than corresponding human-written datasets. We generate 154 datasets and discover new cases of inverse scaling where LMs get worse with size. Larger LMs repeat back a dialog user's preferred answer ("sycophancy") and express greater desire to pursue concerning goals like resource acquisition and goal preservation. We also find some of the first examples of inverse scaling in RL from Human Feedback (RLHF), where more RLHF makes LMs worse. For example, RLHF makes LMs express stronger political views (on gun rights and immigration) and a greater desire to avoid shut down. Overall, LM-written evaluations are high-quality and let us quickly discover many novel LM behaviors.

The number of international benchmarking competitions is steadily increasing in various fields of machine learning (ML) research and practice. So far, however, little is known about the common practice as well as bottlenecks faced by the community in tackling the research questions posed. To shed light on the status quo of algorithm development in the specific field of biomedical imaging analysis, we designed an international survey that was issued to all participants of challenges conducted in conjunction with the IEEE ISBI 2021 and MICCAI 2021 conferences (80 competitions in total). The survey covered participants' expertise and working environments, their chosen strategies, as well as algorithm characteristics. A median of 72% challenge participants took part in the survey. According to our results, knowledge exchange was the primary incentive (70%) for participation, while the reception of prize money played only a minor role (16%). While a median of 80 working hours was spent on method development, a large portion of participants stated that they did not have enough time for method development (32%). 25% perceived the infrastructure to be a bottleneck. Overall, 94% of all solutions were deep learning-based. Of these, 84% were based on standard architectures. 43% of the respondents reported that the data samples (e.g., images) were too large to be processed at once. This was most commonly addressed by patch-based training (69%), downsampling (37%), and solving 3D analysis tasks as a series of 2D tasks. K-fold cross-validation on the training set was performed by only 37% of the participants and only 50% of the participants performed ensembling based on multiple identical models (61%) or heterogeneous models (39%). 48% of the respondents applied postprocessing steps.

As AI systems become more capable, we would like to enlist their help to supervise other AIs. We experiment with methods for training a harmless AI assistant through self-improvement, without any human labels identifying harmful outputs. The only human oversight is provided through a list of rules or principles, and so we refer to the method as 'Constitutional AI'. The process involves both a supervised learning and a reinforcement learning phase. In the supervised phase we sample from an initial model, then generate self-critiques and revisions, and then finetune the original model on revised responses. In the RL phase, we sample from the finetuned model, use a model to evaluate which of the two samples is better, and then train a preference model from this dataset of AI preferences. We then train with RL using the preference model as the reward signal, i.e. we use 'RL from AI Feedback' (RLAIF). As a result we are able to train a harmless but non-evasive AI assistant that engages with harmful queries by explaining its objections to them. Both the SL and RL methods can leverage chain-of-thought style reasoning to improve the human-judged performance and transparency of AI decision making. These methods make it possible to control AI behavior more precisely and with far fewer human labels.

Visual localization is the task of estimating camera pose in a known scene, which is an essential problem in robotics and computer vision. However, long-term visual localization is still a challenge due to the environmental appearance changes caused by lighting and seasons. While techniques exist to address appearance changes using neural networks, these methods typically require ground-truth pose information to generate accurate image correspondences or act as a supervisory signal during training. In this paper, we present a novel self-supervised feature learning framework for metric visual localization. We use a sequence-based image matching algorithm across different sequences of images (i.e., experiences) to generate image correspondences without ground-truth labels. We can then sample image pairs to train a deep neural network that learns sparse features with associated descriptors and scores without ground-truth pose supervision. The learned features can be used together with a classical pose estimator for visual stereo localization. We validate the learned features by integrating with an existing Visual Teach & Repeat pipeline to perform closed-loop localization experiments under different lighting conditions for a total of 22.4 km.

The NASA Astrophysics Data System (ADS) is an essential tool for researchers that allows them to explore the astronomy and astrophysics scientific literature, but it has yet to exploit recent advances in natural language processing. At ADASS 2021, we introduced astroBERT, a machine learning language model tailored to the text used in astronomy papers in ADS. In this work we: - announce the first public release of the astroBERT language model; - show how astroBERT improves over existing public language models on astrophysics specific tasks; - and detail how ADS plans to harness the unique structure of scientific papers, the citation graph and citation context, to further improve astroBERT.

Developing safe and useful general-purpose AI systems will require us to make progress on scalable oversight: the problem of supervising systems that potentially outperform us on most skills relevant to the task at hand. Empirical work on this problem is not straightforward, since we do not yet have systems that broadly exceed our abilities. This paper discusses one of the major ways we think about this problem, with a focus on how to turn it into one that can be productively studied empirically. We first present an experimental design centered on choosing tasks for which human specialists succeed but unaided humans and current general AI systems fail. We then present a proof-of-concept experiment following meant to demonstrate a key feature of this experimental design and show its viability with two question-answering tasks: MMLU and time-limited QuALITY. On these tasks, we find that human participants who interact with an unreliable large-language-model dialog assistant through chat -- a trivial baseline strategy for scalable oversight -- substantially outperform both the model alone and their own unaided performance. These results are an encouraging sign that scalable oversight will be tractable to study with present models and bolster recent findings that large language models can productively assist humans with difficult tasks.

在整个计算科学中，越来越需要利用原始计算马力的持续改进，通过对蛮力的尺度锻炼的尺度增加，以增加网状元素数量的增加。例如，如果不考虑分子水平的相互作用，就不可能对纳米多孔介质的转运进行定量预测，即从紧密的页岩地层提取至关重要的碳氢化合物。同样，惯性限制融合模拟依赖于数值扩散来模拟分子效应，例如非本地转运和混合，而无需真正考虑分子相互作用。考虑到这两个不同的应用程序，我们开发了一种新颖的功能，该功能使用主动学习方法来优化局部细尺度模拟的使用来告知粗尺度流体动力学。我们的方法解决了三个挑战：预测连续性粗尺度轨迹，以推测执行新的精细分子动力学计算，动态地更新细度计算中的粗尺度，并量化神经网络模型中的不确定性。

频率调制连续波（FMCW）LIDAR是一种最近新兴的技术，可通过多普勒效应效率进行每次返回的瞬时相对径向速度测量。在这封信中，我们使用这些多普勒速度测量值从FMCW激光雷达（FMCW Lidar）介绍了第一个连续的一次性绕线算法算法，以帮助几何变性环境中的探测率。我们应用现有的连续时间框架，该框架使用高斯工艺回归有效地估算车辆轨迹，以补偿由于任何机械驱动的激光雷达（FMCW和非FMCW）的扫描性质而引起的运动失真。我们在几个现实世界数据集上评估了我们提出的算法，包括我们收集的公开可用数据集和数据集。我们的算法优于也使用多普勒速度测量值的唯一现有方法，我们研究了包括此额外信息在内的困难条件，可大大提高性能。我们还证明了在标称条件下使用多普勒速度测量值的情况下，仅在有和不使用多普勒速度测量的情况下，仅激光射击的前进量的最新性能。该项目的代码可以在以下网址找到：https：//github.com/utiasasrl/steam_icp。

光学相干断层扫描（OCT）是一种非侵入性技术，可在微米分辨率中捕获视网膜的横截面区域。它已被广泛用作辅助成像参考，以检测与眼睛有关的病理学并预测疾病特征的纵向进展。视网膜层分割是至关重要的特征提取技术之一，其中视网膜层厚度的变化和由于液体的存在而引起的视网膜层变形高度相关，与多种流行性眼部疾病（如糖尿病性视网膜病）和年龄相关的黄斑疾病高度相关。变性（AMD）。但是，这些图像是从具有不同强度分布或换句话说的不同设备中获取的，属于不同的成像域。本文提出了一种分割引导的域适应方法，以将来自多个设备的图像调整为单个图像域，其中可用的最先进的预训练模型可用。它避免了即将推出的新数据集的手动标签的时间消耗以及现有网络的重新培训。网络的语义一致性和全球特征一致性将最大程度地减少许多研究人员报告的幻觉效果，这些效应对周期矛盾的生成对抗网络（Cyclegan）体系结构。

语言模型既展示了定量的改进，又展示了新的定性功能，随着规模的增加。尽管它们具有潜在的变革性影响，但这些新能力的特征却很差。为了为未来的研究提供信息，为破坏性的新模型能力做准备，并改善社会有害的效果，至关重要的是，我们必须了解目前和近乎未来的能力和语言模型的局限性。为了应对这一挑战，我们介绍了超越模仿游戏基准（Big Bench）。 Big Bench目前由204个任务组成，由132家机构的442位作者贡献。任务主题是多样的，从语言学，儿童发展，数学，常识性推理，生物学，物理学，社会偏见，软件开发等等。 Big-Bench专注于被认为超出当前语言模型的功能的任务。我们评估了OpenAI的GPT型号，Google内部密集变压器体系结构和大型基础上的开关稀疏变压器的行为，跨越了数百万到数十亿个参数。此外，一个人类专家评估者团队执行了所有任务，以提供强大的基准。研究结果包括：模型性能和校准都随规模改善，但绝对的术语（以及与评估者的性能相比）；在模型类中的性能非常相似，尽管带有稀疏性。逐渐和预测的任务通常涉及大量知识或记忆成分，而在临界规模上表现出“突破性”行为的任务通常涉及多个步骤或组成部分或脆性指标；社交偏见通常会随着含糊不清的环境而随着规模而增加，但这可以通过提示来改善。