In this paper, we study the harmlessness alignment problem of multimodal large language models (MLLMs). We conduct a systematic empirical analysis of the harmlessness performance of representative MLLMs and reveal that the image input poses the alignment vulnerability of MLLMs. Inspired by this, we propose a novel jailbreak method named HADES, which hides and amplifies the harmfulness of the malicious intent within the text input, using meticulously crafted images. Experimental results show that HADES can effectively jailbreak existing MLLMs, which achieves an average Attack Success Rate (ASR) of 90.26% for LLaVA-1.5 and 71.60% for Gemini Pro Vision. Our code and data are available at https://github.com/RUCAIBox/HADES. Warning: this paper contains example data that may be offensive.

It is often used to justify the effectiveness of machine learning algorithms in high-dimensional settings, since the curse of dimensionality can be circumvented if the data concentrates on a lowdimensional manifold. It is, however, evident that several low-dimensional (and hence, visualisable) datasets do not satisfy the Manifold Hypothesis. Instead, such data can have singularities -- points at which the local geometry does not resemble n-dimensional Euclidean space for any n. Prime examples of singular loci of datasets include branching points in neurons and cosmic filaments. Furthermore, standard image datasets (such as MNIST and CIFAR-10) are known to have non-constant intrinsic dimension [17], whereas a connected manifold must possess the same intrinsic dimension throughout. Whenever such non-manifold behaviour within datasets is of interest, it becomes natural to wonder whether it can be accurately and automatically identified. Particularly in large, high-dimensional datasets where visual inspection is impossible, we seek tools to identify and locate singularities within datasets. Our focus here is on unsupervised singularity detection, where one has recourse neither to a plethora of training data, nor the opportunity to regenerate samples along an unknown probability measure.

This paper introduces HADES, a novel tool for automatic comparative documents with similar structures. HADES is designed to streamline the work of professionals dealing with large volumes of documents, such as policy documents, legal acts, and scientific papers. The tool employs a multi-step pipeline that begins with processing PDF documents using topic modeling, summarization, and analysis of the most important words for each topic. The process concludes with an interactive web app with visualizations that facilitate the comparison of the documents. HADES has the potential to significantly improve the productivity of professionals dealing with high volumes of documents, reducing the time and effort required to complete tasks related to comparative document analysis. Our package is publically available on GitHub.

Prompt and accurate detection of system anomalies is essential to ensure the reliability of software systems. Unlike manual efforts that exploit all available run-time information, existing approaches usually leverage only a single type of monitoring data (often logs or metrics) or fail to make effective use of the joint information among different types of data. Consequently, many false predictions occur. To better understand the manifestations of system anomalies, we conduct a systematical study on a large amount of heterogeneous data, i.e., logs and metrics. Our study demonstrates that logs and metrics can manifest system anomalies collaboratively and complementarily, and neither of them only is sufficient. Thus, integrating heterogeneous data can help recover the complete picture of a system's health status. In this context, we propose Hades, the first end-to-end semi-supervised approach to effectively identify system anomalies based on heterogeneous data. Our approach employs a hierarchical architecture to learn a global representation of the system status by fusing log semantics and metric patterns. It captures discriminative features and meaningful interactions from heterogeneous data via a cross-modal attention module, trained in a semi-supervised manner. We evaluate Hades extensively on large-scale simulated data and datasets from Huawei Cloud. The experimental results present the effectiveness of our model in detecting system anomalies. We also release the code and the annotated dataset for replication and future research.

Last week, like many Americans, I traveled home for Thanksgiving. I chose not to bring my Switch with me because, well, I've sort of moved on. It's old and underpowered compared to the PlayStation 5 and Xbox Series X, and, besides, the last couple years of being a homebody (thanks to lockdowns and remote work) have rendered its portability meaningless to me. And then I actually caught (a very mild case of) COVID and was stranded in Kansas for an extra week with no video games. Readers, I've never appreciated the Switch more than I did when it would have saved me from a week of extreme boredom.

So how do you categorize a beast like gaming on the PC? With decades of titles to pluck from (and the first port of call for most indie titles, too), there's so much to choose from. Gaming on your PC adds the benefits of (nearly always flawless) backward compatibility and console-beating graphical performance -- if you've got the coin for it. The whole idea of what a PC is and where you can play it is shifting, too, with the rise of handheld "consolized" PCs like the Steam Deck. We've tried to be broad with our recommendations here on purpose – there are so many great games out there for your PC, consider these some starting points.

Welcome back to Pushing Buttons, folks. In case you're wondering where I've been, I was on secondment from video games for a week covering Glastonbury. Thankfully, a decade-plus of E3 coverage prepared me well for the fragrant crowds and inevitable liveblogging tech issues. Thank you to our ever-brilliant games correspondent Keith Stuart for covering for me while I screwed my head back on after the festival. Pride events took place in London last weekend, and among the million people lining the streets for the event's 50th-anniversary were parade contingents from PlayStation, Microsoft and Square Enix, among other game publishers and developers.

Welcome to Pushing Buttons, the Guardian's gaming newsletter. If you'd like to receive it in your inbox every week, just pop your email in below – and check your inbox (and spam) for the confirmation email. This week marks a truly important video game anniversary: it is 50 years since Nolan Bushnell and Ted Dabney incorporated Atari Inc, the company that laid the foundations for the video games industry. There have been many appraisals of the company and its landmark achievements in the games press over the past few days – from the arrival of a Pong machine in Andy Capp's Tavern in Sunnyvale, California, in 1972, through classic titles such as Breakout, Asteroids and Missile Commands, to the iconic home consoles. But one element that often gets overlooked in these nostalgic reveries is the way in which Atari taught the first generation of electronic gamers how to think symbolically.

Months after the finalists were announced, a video game has won a Hugo Award for the first time. Gizmodo reports Supergiant Games' Hades has won the one-time award for Best Video Game, beating out the likes of Animal Crossing and The Last of Us: Part II (among others). Creative director Greg Kasavin gave a virtual acceptance speech saying he was "grateful" organizers were recognizing work in the game industry. Hades has been available since September 2020, shortly before the Hugo Awards revealed plans for a game category in November that year. It came to PlayStation and Xbox consoles this August, however. The title has generally received a warm reception through fast-paced roguelike gameplay, a solid narrative element and mechanics that reward repeat playthroughs.

Just five years ago, Nintendo was at a crossroads. The Wii U was languishing well in third place in the console wars and, after considerable pressure, the company was making its first tentative steps into mobile gaming with Miitomo and Super Mario Run. Fast-forward to today: The Switch is likely on the way to becoming the company's best-selling "home console" ever, and seven Switch games have outsold the Wii U console. However, the Switch's online store isn't the easiest to navigate, so this guide aims to help the uninitiated start their journey on the right foot. These are the games you should own -- for now.