Deep ensembles (DE) have been successful in improving model performance by learning diverse members via the stochasticity of random initialization. While recent works have attempted to promote further diversity in DE via hyperparameters or regularizing loss functions, these methods primarily still rely on a stochastic approach to explore the hypothesis space. In this work, we present Multi-Symmetry Ensembles (MSE), a framework for constructing diverse ensembles by capturing the multiplicity of hypotheses along symmetry axes, which explore the hypothesis space beyond stochastic perturbations of model weights and hyperparameters. We leverage recent advances in contrastive representation learning to create models that separately capture opposing hypotheses of invariant and equivariant functional classes and present a simple ensembling approach to efficiently combine appropriate hypotheses for a given task. We show that MSE effectively captures the multiplicity of conflicting hypotheses that is often required in large, diverse datasets like ImageNet. As a result of their inherent diversity, MSE improves classification performance, uncertainty quantification, and generalization across a series of transfer tasks.

Distribution shift is a major source of failure for machine learning models. However, evaluating model reliability under distribution shift can be challenging, especially since it may be difficult to acquire counterfactual examples that exhibit a specified shift. In this work, we introduce the notion of a dataset interface: a framework that, given an input dataset and a user-specified shift, returns instances from that input distribution that exhibit the desired shift. We study a number of natural implementations for such an interface, and find that they often introduce confounding shifts that complicate model evaluation. Motivated by this, we propose a dataset interface implementation that leverages Textual Inversion to tailor generation to the input distribution. We then demonstrate how applying this dataset interface to the ImageNet dataset enables studying model behavior across a diverse array of distribution shifts, including variations in background, lighting, and attributes of the objects. Code available at https://github.com/MadryLab/dataset-interfaces.

Decentralized Autonomous Organization (DAO) provides a decentralized governance solution through blockchain, where decision-making process relies on on-chain voting and follows majority rule. This paper focuses on MakerDAO, and we find three voter coalitions after applying clustering algorithm to voting history. The emergence of a dominant voter coalition is a signal of governance centralization in DAO, and voter coalitions have complicated influence on Maker protocol, which is governed by MakerDAO. This paper presents empirical evidence of multicoalition democracy in DAO and further contributes to the contemporary debate on whether decentralized governance is possible.

Sensors are crucial for autonomous operation in robotic vehicles (RV). Unfortunately, RV sensors can be compromised by physical attacks such as tampering or spoofing, leading to a crash. In this paper, we present DeLorean, a modelfree recovery framework for recovering autonomous RVs from sensor deception attacks (SDA). DeLorean is designed to recover RVs even from a strong SDA in which the adversary targets multiple heterogeneous sensors simultaneously (even all the sensors). Under SDAs, DeLorean inspects the attack induced errors, identifies the targeted sensors, and prevents the erroneous sensor inputs from being used to derive actuator signals. DeLorean then replays historic state information in the RV's feedback control loop for a temporary mitigation and recovers the RV from SDA. Our evaluation on four real and two simulated RVs shows that DeLorean can recover RVs from SDAs, and ensure mission success in 90.7% of the cases on average.

Language Models (LMs) have proven to be useful in various downstream applications, such as summarisation, translation, question answering and text classification. LMs are becoming increasingly important tools in Artificial Intelligence, because of the vast quantity of information they can store. In this work, we present ProP (Prompting as Probing), which utilizes GPT-3, a large Language Model originally proposed by OpenAI in 2020, to perform the task of Knowledge Base Construction (KBC). ProP implements a multi-step approach that combines a variety of prompting techniques to achieve this. Our results show that manual prompt curation is essential, that the LM must be encouraged to give answer sets of variable lengths, in particular including empty answer sets, that true/false questions are a useful device to increase precision on suggestions generated by the LM, that the size of the LM is a crucial factor, and that a dictionary of entity aliases improves the LM score. Our evaluation study indicates that these proposed techniques can substantially enhance the quality of the final predictions: ProP won track 2 of the LM-KBC competition, outperforming the baseline by 36.4 percentage points.

For population studies or for the training of complex machine learning models, it is often required to gather data from different actors. In these applications, summation is an important primitive: for computing means, counts or mini-batch gradients. In many cases, the data is privacysensitive and therefore cannot be collected on a central server. Hence the summation needs to be performed in a distributed and privacy-preserving way. Existing solutions for distributed summation with computational privacy guarantees make trust or connection assumptions -- e.g., the existence of a trusted server or peer-to-peer connections between clients -- that might not be fulfilled in real world settings. Motivated by these challenges, we propose Secure Summation via Subset Sums (S5), a method for distributed summation that works in the presence of a malicious server and only two honest clients, and without the need for peer-to-peer connections between clients. S5 adds zero-sum noise to clients' messages and shuffles them before sending them to the aggregating server. Our main contribution is a proof that this scheme yields a computational privacy guarantee based on the multidimensional subset sum problem. Our analysis of this problem may be of independent interest for other privacy and cryptography applications.

Neil Thomas wished he could have been awake during the operation to remove a 6cm cancerous tumour from his colon. He was one of the first people to go under the scalpel of University Hospital of Wales's new robotic systems in June 2022. Thomas's surgeon, James Ansell, would once have stooped over his patient's body to perform the operation. Instead, he stood behind a console on another side of the theatre wearing 3D glasses. His hands grasped two joysticks, which controlled the four robotic arms that huddled around Thomas's unconscious body.

Astronaut Sally Ride became the first American woman in space on this day in history, June 18, 1983. Born on May 26, 1951, in Los Angeles, Ride earned bachelor's degrees in English and physics from Stanford University in California before staying at Stanford and earning a PhD in physics in 1978. Shortly before earning her doctorate, Ride saw an ad for a newspaper that piqued her interest. NASA was recruiting for astronauts -- and, for the first time, the agency would include women in its astronaut class. "Over 8,000 men and women applied to the space program that year. Of the 35 individuals accepted, six were women, and I was one of them. This was in January 1978," said Ride in quotes listed on a tribute page on NASA's website.

Academic literature does not give much guidance on how to build the best possible customer-facing summarization system from existing research components. Here we present analyses to inform the selection of a system backbone from popular models; we find that in both automatic and human evaluation, BART performs better than PEGASUS and T5. We also find that when applied cross-domain, summarizers exhibit considerably worse performance. At the same time, a system fine-tuned on heterogeneous domains performs well on all domains and will be most suitable for a broad-domain summarizer. Our work highlights the need for heterogeneous domain summarization benchmarks. We find considerable variation in system output that can be captured only with human evaluation and are thus unlikely to be reflected in standard leaderboards with only automatic evaluation.

Rumor spreaders are increasingly utilizing multimedia content to attract the attention and trust of news consumers. Though quite a few rumor detection models have exploited the multi-modal data, they seldom consider the inconsistent semantics between images and texts, and rarely spot the inconsistency among the post contents and background knowledge. In addition, they commonly assume the completeness of multiple modalities and thus are incapable of handling handle missing modalities in real-life scenarios. Motivated by the intuition that rumors in social media are more likely to have inconsistent semantics, a novel Knowledge-guided Dual-consistency Network is proposed to detect rumors with multimedia contents. It uses two consistency detection subnetworks to capture the inconsistency at the cross-modal level and the content-knowledge level simultaneously. It also enables robust multi-modal representation learning under different missing visual modality conditions, using a special token to discriminate between posts with visual modality and posts without visual modality. Extensive experiments on three public real-world multimedia datasets demonstrate that our framework can outperform the state-of-the-art baselines under both complete and incomplete modality conditions. Our codes are available at https://github.com/MengzSun/KDCN.